UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Expert-novice differences in knowledge structures of action Vickers, Joan N. 1983

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1984_A2 V53.pdf [ 7.1MB ]
Metadata
JSON: 831-1.0076984.json
JSON-LD: 831-1.0076984-ld.json
RDF/XML (Pretty): 831-1.0076984-rdf.xml
RDF/JSON: 831-1.0076984-rdf.json
Turtle: 831-1.0076984-turtle.txt
N-Triples: 831-1.0076984-rdf-ntriples.txt
Original Record: 831-1.0076984-source.json
Full Text
831-1.0076984-fulltext.txt
Citation
831-1.0076984.ris

Full Text

EXPERT-NOVICE DIFFERENCES IN KNOWLEDGE STRUCTURES OF ACTION by JOAN N. VICKERS B.P.E., University of New Brunswick, 1966 M.Sc, University of Calgary, 1976 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION (Inter-Departmental) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December, 1983 Q Joan N. Vickers, 1983 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying o f t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head o f my department o r by h i s or her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department of The U n i v e r s i t y of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 i i Abstract A s e r i e s of three studies was conducted to determine differences between three groups of gymnasts i n a perception - eye movement study, i n a problem solving - resequencing study and i n an introspection study. T h i r t y subjects were selected and assigned to three groups based on t h e i r expertise i n gymnastics. Groups one and two, e l i t e and intermediate gymnasts, were selected from a n a t i o n a l l y ranked gymnastic club. Group three, the novice gymnasts, were members of a championship soccer team. A l l subjects were female, average age 13.2 years. Differences between the groups were consistent across the three studies, lending support to the notion that as s k i l l develops there are changes i n the i n t e r n a l representation or knowledge structure of the athlete that are detectable by means of perception, problem solving and other cognitive tasks. In the perception study, the eye movements of the subjects were taken as they viewed slow motion s l i d e presentations of world c l a s s gymnasts performing basic compulsory moves. In the problem solving study, the subjects resequenced sets of photos taken from the s l i d e s e r i e s used i n the f i r s t experiment. The t h i r d study was an introspection study i n which each subject was shown the i n t a c t photographic sequences and asked to v e r b a l l y i d e n t i f y where i n the i i i movement they concentrated their attention, i f they were performing the movement in competition. In the total sequence analysis of eye movements, differences between the e l i t e and intermediate/novice gymnasts were confirmed by multivariate analyses in two of the six sequences and by univariate analyses of separate body segments in four of the six sequences. MANOVA also detected differences between the intermediates and novices in three of the six sequences and by univariate analysis of separate body segments in five of the six sequences. Differences were observed between the groups in how they observed the head, hips, and offbody factors. In the phase analysis of one sequence, the beam handstand, differences in eye movements were found between the groups relative to whether the beginning, mid or end phases of the sequence were being analysed. The results from the resequencing study confirmed that the e l i t e gymnasts were both faster and more accurate than the intermediates in reconstructing the gymnastic sequences; the intermediates in turn were both faster and committed fewer errors than the novices. It was also found that the more d i f f i c u l t the gymnastic sequence, the more time needed to reconstruct and the greater the number of errors. The f i n a l two aspects of the research also found significant differences between the groups. F i r s t , eye fixations to specific body segments were predictive of resequencing time scores. A iv multiple regression analysis showed that 43.3% of the variance in the resequencing time scores was accounted for by eye fixation patterns. Second, across the six sequences, the introspection reports were congruent with the eye movement data (70.4%) but the groups differed in their selection of body segments. A log linear model analysis of the count data showed that the e l i t e gymnasts were lower body oriented (63.5%), the intermediates were upper body oriented (73.9%), while the novices were 60% upper and 40% lower. In summary, the results were clear in showing that when asked to observe gymnastic s k i l l s or solve problems with or introspect about stimulus material from gymnastics the groups differed consistently. These differences, of mutifaceted nature, were taken as contributing to the identification of relevant attributes of the knowledge structure of gymnasts. Implications of these findings were discussed in terms of their applicability to research on s k i l l acquisition, and their potential role as diagnostic and prescriptive factors in teaching and coaching. TABLE OF CONTENTS I. STATEMENT OF THE RESEARCH PROBLEM A. Introduction 1 B. Knowledge structures of a c t i o n 2 Knowledge structures of ac t i o n defined 4 Expert and novice knowledge structures 5 C. A conceptual framework f or exploring knowledge structures of action 7 Three cognitive studies 8 S p e c i f i c research questions and expectations . . . 9 D. V i s u a l perception as the "minds eye" of act i o n . 10 Perceiving pictures of human movement 12 Perceptual i n v a r i a n t s : exploring information i n dynamic events 16 I I . METHOD A. Eye movement measures: A method for detecting differences i n knowledge structures of action . 22 B. The resequencing task: Conceptually driven and data driven perceptual systems 27 C. The intr o s p e c t i o n task 28 D. Subjects and design 29 v i E. The eye movement study Apparatus 30 Stimuli 32 Procedure 34 F. The resequencing study Procedure 34 G. The introspection study .• 36 H. Eye movement and intr o s p e c t i o n coding system . . 37 Rationale f o r data combination 39 I I I . RESULTS A. The eye movement study: T o t a l sequence and phase analysis 41 B. Analysis of the resequencing performance . . . . 42 C. The r e l a t i o n s h i p between the introspection reports and eye movements 50 D. The congruence between the intr o s p e c t i o n reports and eye movements 61 IV. SUMMARY OF RESULTS, DISCUSSION AND CONCLUSION A. Summary of r e s u l t s 65 B. Discussion 69 C. Uniqueness and implications of the research . . 74 D. Conclusion 79 REFERENCES 8 0 TABLES 9 1 FIGURES 1 U APPENDICES 1 2 6 v i i i LIST OF TABLES TABLE 1 Three l e v e l s of coding as applied eye movement data i n s i x movement sequences i n gymnastics. . 91 2 Eye movements t o t a l sequence analysis f o r forward walkover. Observed mean 3 x 2 contract e f f e c t s f o r six body segments. . -. 92 3 Eye movements t o t a l sequence analysis f o r forward handspring. Observed mean 3 x 2 contract e f f e c t s for s i x body segments . . . . . . . . . 93 4. Eye movements t o t a l sequence analysis for uneven bars. Observed mean 3 x 2 contrast e f f e c t s f o r s i x body segments . . <. 94 5 Eye movements t o t a l sequence analysis for beam back walkover. Observed mean 3 x 2 contrast e f f e c t s for s i x body segments 95 Eye movements t o t a l sequence analysis f o r beam handstand. Observed mean 3 x 2 contrast e f f e c t s f o r s i x body segments . . 96 Eye movements t o t a l sequence analysis f o r va u l t . Observed mean 3 x 2 contrast e f f e c t s f o r s i x body segments 97 Eye movements t o t a l sequence an a l y s i s : Description of constant contrast e f f e c t s by s i x body segments. P r o f i l e s of body segments i r r e s p e c t i v e of expertise and f i x a t i o n l e v e l s 98 Eye movements t o t a l sequence an a l y s i s : Testing of two expertise l e v e l contrasts e f f e c t s by s i x body segments. Expertise x body segment i n t e r a c t i o n i r r e s p e c t i v e f i x a t i o n l e v e l s 99 Eye movements t o t a l sequence analysis: Summary of differences i n f i x a t i o n s to body segments i n s i x sequences i r r e s p e c t i v e of f i x a t i o n l e v e l . . . . 100 Eye movements t o t a l sequence an a l y s i s : Testing of f i r s t vs. second glance contrast e f f e c t s by s i x body segments. F i r s t vs. second glance x body segment i n t e r a c t i o n i r r e s p e c t i v e of expertise Eye movements t o t a l sequence analysis of variance: Testing of 2 expertise l e v e l contrasts e f f e c t s x f i r s t vs. second glance by s i x body segments, expertise x f i x a t i o n : F i r s t vs. second glance x body segment i n t e r a c t i o n . 102 Eye movement phase an a l y s i s : Mean number of f i x a t i o n s to s i x body segments f o r beamhand 103 Eye movement phase a n a l y s i s : Observed means of polynomial component by s i x body segments. Trend across 4 phases by 3 expertise l e v e l s 104 Eye movement phase a n a l y s i s : Observed 3 expertise l e v e l contrasts and 4 phase polynomial trend e f f e c t s by si x body segments .105 Eye movement phase analysis of the beamhand: Testing of 4 phase polynomial trend components e f f e c t s by s i x body segments for grand mean of expertise l e v e l s phase and body segment i n t e r a c t i o n i r r e s p e c t i v e of expertise l e v e l s 106 Eye movements phase analysis of the beamhand: Testing of 2 expertise l e v e l contrast x phase polynomial trend components e f f e c t s by six body segments (expertise x phase x body segments interaction) 107 l e v e l 101 X 18 Analysis of the resequencing task: mean time and errors per s l i d e for s i x sequences. 108 19 Analysis of the resequencing task: Observed 3 expertise l e v e l Contrasts and 6 Sequence Contrasts E f f e c t s . 109 20 Analysis of the resequencing task: Testing of 3 expertise l e v e l contrasts x 6 sequence contrasts e f f e c t s . 110 21 Means and i n t e r c o r r e l a t i o n matrix of resequencing task. Completion time and s i x selected eye f i x a t i o n s scores (beamhand) (n=30) I l l 22 Frequency contingency table of intr o s p e c t i o n responses congruent with eye f i x a t i o n s to s i x body segments by expertise l e v e l s and s i x sequences 112 23 Log l i n e a r analysis o f a 3 x 6 x 2 x 2 chi-squares contingency table (Table 27) by orthogonal sets of contrasts ,113 x i LIST OF FIGURES Figure 1 A conceptual framework for exploring knowledge structures of action 114 2. Two examples of eye movement data. In photograph 2a, the shoulder i s fixated, code 02; in photograph 2b, movement space i s shown, code 18 116 3 The beam handstand sequence showing twelve of the twenty-six slides used in the eye movement study 118 4 Mean fixations to six body segments at four phases of the beam handstand sequence 120 5 Mean time per photograph in resequencing six movement sequences in gymnastics 122 6 Mean errors per photograph in resequencing six movement sequences in gymnastics 124 x i i ACKNOWLEDGEMENT I would l i k e to thank my supervisor, Soeng Soo Lee, for h i s famous attention to d e t a i l , h i s even more famous tenacity and h i s constant support throughout the preparation of t h i s d i s s e r t a t i o n . My thanks also to Stan Coren, who introduced me to eye movements research, and whose willingness to explore the r e l a t i o n s h i p between act i o n and v i s i o n was much appreciated. My thanks also to Ian Franks, f o r h i s in s i g h t s i n t o motor s k i l l a c q u i s i t i o n and to Ann Treisman and Daniel Kahneman who were the f i r s t to introduce me to the f i e l d of cognition. The best of luck and thank-you to the F l i c k a gymnastic club athletes and t h e i r coach Mike Vossen, and to Calvin L a i who was so h e l p f u l i n w r i t i n g some of the computer programs. I salute my sons Jamie and Robert, who never once complained or questioned the time spent on t h i s p r o j ect. F i n a l l y , and most importantly, my deepest thanks and fondest regards to Bob Morford, who's steadfast support has made a l l the difference i n the world. 1. I. Statement of the Research Problem A. Introduction He was the ultimate in what you would expect in the gifted a t h l e t e — s i x foot two, one hundred and ninety pounds, a member of his National team and a medalist as well. And yet, he could not perform this new activity, no matter how hard he tried. To make matters worse, his GPA was 3.9, he played in the Symphony orchestra of the large city where he lived, and he willingly practised and practised. He seemed to represent the one indisputable case where inability could not be attributed to physical size, strength, coordination, intelligence or motivation. In his case, I reasoned, he had taken a superbly capable physical body and constrained i t by asking i t to perform in a way that was known to be ineffective; he seemed to have the wrong idea, he perceived the strokes in the game in a way that was biomechanically and conceptually inaccurate. No amount of explanation, analysis or feedback could change the concepts he carried in his mind. Now i f i t were somehow possible to get inside his head, to read his mind as i t were, would we not have available a unique and new type of information that would be useful in teaching and coaching, in instructional design and a host of other areas? Such was the question that lead to the following studies. The inquiry i s couched under an a l l encompassing and 2. admittedly l a r g e l y undefined term c a l l e d knowledge structures of a c t i o n . B. Knowledge structures of action The study of the nature of knowledge and i t s organization within the human system has resulted i n the evolution of a host of terms such as concepts (Rumelhart, Lindsay, and Norman, 1972), schemes, schema or schemata ( B a r t l e t t , 1932; Rumelhart and Ortony, 1977), s c r i p t s (Shank and Abelson, 1977), networks (Wickelgren, 1981; Anderson, 1980), frames (Minsky, 1975), and knowledge structures (Glaser, 1976; Simon, 1980; Chi and Glaser, 1980). Despite the d i v e r s i t y of t h e i r o r i g i n s , a l l of these terms have one idea i n common. They a l l recognize that knowledge i s stored not merely as an accumulative record of sensory information, but rather, i t i s stored possessing information content and organizational structure and innumerable processes for using such information e f f e c t i v e l y . The knowledge that athletes develop has not been studied to any great extent as concepts or as knowledge structures (Newell and Barclay, 1982). However, athletes do possess a complex network of information composed of s k i l l s , s u b s k i l l s , t a c t i c s , s t r a t e g i e s , r u l e s , customs, and t r a d i t i o n s , as well as a host of processes for using t h i s information. The question of how knowledge may be organized i n action contexts has evolved a number of theories that 3. have attempted to explain the r e l a t i o n s h i p between the so-called higher order or cognitive aspects of p h y s i c a l performance and the lower or motor u n i t l e v e l s (Berstein, 1967, p.37; Pew, 1979; Schmidt, 1975; Glencross, 1978; Turvey, 1977a; G a l l i s t e l , 1982, pp. 210-228; Reed, 1982). Bernstein (1967) and Turvey (1977a) have suggested that during the a c q u i s i t i o n of s k i l l , athletes search for an "optimal organisation" that serves to guide t h e i r subsequent actions. This search i s described as a gaining of c o n t r o l over the "degrees of freedom" inherent i n a movement problem. Every p h y s i c a l s k i l l i s proposed to have degrees of freedom at both the higher and lower l e v e l s that must be c o n t r o l l e d before a s k i l l f u l act eventually emerges. Such v a r i a t i o n over a range of l i k e l y or possible movement outcomes occurs because, i n the early stages of s k i l l development, the executive or higher centres are unable to co n t r o l , indeed are incapable of c o n t r o l l i n g , the muscle units i n d i v i d u a l l y . However, as s k i l l develops, human action i s organized i n such a way that increasingly higher l e v e l control i s exercised over f u n c t i o n a l l y larger groups of muscles. The primary purpose of the higher centres therefore i s to c o n t r o l "the modes of i n t e r a c t i o n of the lower centres" such that over time there evolves "co-ordinated movements" that are i n i t i a t e d by r e l a t i v e l y simple executive i n s t r u c t i o n s which sets the e n t i r e segmental apparatus i n motion. With the a c q u i s i t i o n of s k i l l , should come changes i n the type of executive control exercised. 4. This was shown by Pew (1966) who found d i f f e r e n t q u a l i t i e s of higher l e v e l c o n t r o l i n a movement problem when subjects were given equal opportunities to learn a compensatory tracking task that involved second order control using movements of the index fingers of both hands. Af t e r equal amounts of p r a c t i s e a few of the subjects evolved a pattern of hand independence that greatly enhanced t h e i r l e v e l of s k i l l ; they had developed a type of executive l e v e l strategy or cognitive control that f u n c t i o n a l l y a l t e r e d and improved the way the muscle units c a r r i e d out the task. The emergence of hand independence has since been acknowledged as a c h a r a c t e r i s t i c of s k i l l development i n piano playing and typing (Shaffer, 1981; Norman, 1981). Knowledge structures of action defined. The term, knowledge structure of action, has been chosen to denote the i n v e s t i g a t i o n of the higher order or cognitive aspects of p h y s i c a l performance. I t i s assumed that every athlete develops an abstract i n t e r n a l representation of an act that i s created i n concert with t h e i r developing muscle system. This abstract representation i s comprised of informational content, structure and process. The term knowledge structure has been selected for use i n an action context because the athletes' organized knowledge about action i s of i n t e r e s t , more than the k i n e t i c or kinematic aspects that accompany t h e i r behavior. The work presented here i s an exploratory excursion with three p r i n c i p a l aims: 1) to determine i f cognitive studies, i n p a r t i c u l a r those r e l y i n g upon v i s u a l perception, are able to detect differences between groups of athletes with d i f f e r e n t s k i l l l e v e l s ; 2) to describe these differences as a t t r i b u t e s of the a t h l e t e s ' knowledge structure; 3) to discuss the r e s u l t s i n terms of t h e i r a p p l i c a b i l i t y to research on s k i l l a c q u i s i t i o n and t h e i r p o t e n t i a l r o l e as diagnostic and p r e s c r i p t i v e factors i n teaching and coaching. Expert and novice knowledge structures. Those who have studied the differences between experts and novices i n f i e l d s such as chess (Chase and Simon, 1973; Chi, Glaser and Rees, 1982), computer programming (McKeithen, Reitman, Rueter and H i r t l e , 1981), radiology (Nodine and Kundell, 1981), teaching (Donald, 1980), cardiology, commodity trading, chemical engineering and law (Johnson, 1979), provide evidence that "the most s t r i k i n g c h a r a c t e r i s t i c of expertise i s that experts have trouble explaining how they do i t . A l o t of what they do appears to be based on a kind of i n t u i t i o n " (Hunt, 1982, p.264). Despite the i l l u s i v e n e s s that seems to accompany expertise, experimental methods have been successful i n describing expert-novices differences i n knowledge, organization and process. A t y p i c a l experiment by Johnson (1979) i n cardiology compared expert 6. diagnosticians with medical students on how they would treat a p a r t i c u l a r case. Johnson's conclusions were consistent with the work of others. Experts, although they often possess the same amount and type of information that others do, have developed i n t r i c a t e and often subtle l i n k s between pieces of information that are more e f f e c t i v e and applicable i n a p a r t i c u l a r s i t u a t i o n . The expert has a "high a l t i t u d e overview...an enormously e f f i c i e n t p i c t u re of what they are t r y i n g to do," whereas the novice often c o l l e c t s too much information, brings too much to bear on a s i t u a t i o n , and often uses i t inappropriately. F i n a l l y , expertise appears to be domain s p e c i f i c and i n i t i a l hopes of i d e n t i f y i n g generic expert a b i l i t i e s independent of the knowledge i n a p a r t i c u l a r f i e l d have been d i f f i c u l t to e s t a b l i s h . The knowledge that athletes develop has not been studied to any great extent for the obvious reason that i t i s d i f f i c u l t to obtain data from subjects engaged i n dynamic and non-verbal tasks. An even greater b a r r i e r i s implied by statements suggesting that with the attainment of s k i l l , i n p a r t i c u l a r the evolution of automaticity, comes an i n a b i l i t y f o r the a t h l e t e to remember, recount or even understand what they do (Norman, 1982). I t often appears an assumption has been made that implies i t i s too d i f f i c u l t , even impossible, to get at the information; i t i s as though, with the attainment of s k i l l , some mysterious cerebral evaporation has occurred that defies detection and study. Our 7. i n a b i l i t y to detect the cognitive changes that occur as performers becomes more s k i l l e d has led to a legitimate f r u s t r a t i o n with current methodologies (Norman, 1982; Whiting, 1980). However, because highly s k i l l e d athletes seem not to understand the processes that lead to t h e i r high l e v e l s of s k i l l does not mean that the knowledge they possess i s not there to be understood. The problem appears to be more one of methodology, of f i n d i n g ways to tap the knowledge athletes possess i n v a l i d and meaningful ways. To t h i s end a conceptual framework i s presented f o r the purposes of exploring knowledge structures of action as they reside i n the athlete. C. A conceptual framework for exploring knowledge structures of  action The research framework i l l u s t r a t e d i n Figure 1 shows the r e l a t i o n s h i p between three groups of gymnasts, three cognitive tasks and the a t t r i b u t e s of a knowledge structure expressed i n terms of units of a n a l y s i s . Insert Figure 1 about here __^5_?r_^rS^__^A The design i s a comparative one, an expert-novice paradigm. Three groups of gymnasts were selected based on t h e i r l e v e l of expertise i n gymnastics. An a p r i o r i assumption i s that the muscle systems 8. of the performers d i f f e r ; unknown i s whether they also d i f f e r i n terms of perception, problem solving and in t r o s p e c t i o n r e l a t i v e to the gymnastic stimulus information presented. Knowledge structures * i n sporting a c t i v i t i e s are proposed to be highly s p e c i f i c with each a c t i v i t y d i s p l a y i n g a uniqueness that sets i t aside from a l l others. For t h i s reason, only the information that defines c e r t a i n basic s k i l l s within gymnastics i s considered. Three cognitive studies. Three i n t e r r e l a t e d studies were designed to test the general hypothesis that an athlete's knowledge structure of t h e i r event d i f f e r s as a function of t h e i r l e v e l of expertise. The f i r s t was a perception study where the eye movements of the subjects were taken as they viewed slow motion s l i d e presentations of world c l a s s gymnasts performing basic compulsory moves. The second was a problem solving study where the subjects resequenced or reconstructed sets of photographs of gymnastic sequences taken from the s l i d e s e r i e s used i n the f i r s t experiment. The t h i r d study was an introspection study i n which each subject was shown the i n t a c t photographic sequences and asked to v e r b a l l y i d e n t i f y where within the movement they would concentrate t h e i r inner focus of attention, were they performing the move i n competition. Taken i n t o t a l , these three studies sought to determine the extent to which d i f f e r e n t types of v i s u a l tasks d i f f e r e n t i a t e d between subjects of varying s k i l l l e v e l s . 9. S p e c i f i c research questions and expectations. In the eye movements study, differences between the groups i n f i x a t i o n s to body segments were explored with two expectations. F i r s t , i t was expected that the groups would d i f f e r i n f i x a t i o n s to body segments. Since there i s l i t t l e relevant information a v a i l a b l e for t h i s study i n e i t h e r the eye movements or the applied motor learning l i t e r a t u r e , no attempt was made to specify d i r e c t i o n a l differences between the groups i n f i x a t i o n s to s p e c i f i c body segments. Secondly, i t was expected that the e l i t e subjects would ex h i b i t fewer f i x a t i o n s to body segments than the intermediates and novices. This expectation was i n agreement with eye movements studies of expert and novice basketball and i c e hockey players and expert and novice judges i n gymnastics (Bard and Fleury, 1976; Bard, Fleury, Carriere and H a l l e , 1980; Bard and Fleury, 1981). In the second study, the v i s u a l resequencing task, the three groups were presented with v i s u a l s t i m u l i consisting of 11 or 12 photographs depicting the gymnastic sequences viewed i n the eye movements phase of the research. The photographs were presented to each subject randomly out-of-order and t h e i r task was to reconstruct each sequence as quickly and as accurately as possible. The e l i t e subjects were expected to be f a s t e r and commit fewer errors than the intermediates and novices; i n turn, the intermediates were expected to be both f a s t e r and more accurate than the novices. In addition, i t was expected that the greater 10. the degree of d i f f i c u l t y of a movement, the greater the l i k e l i h o o d of s i g n i f i c a n t differences between the groups. Floor moves were c l a s s i f i e d as having a lower degree of d i f f i c u l t y than equipment e v e n t s — t h e beam, vault and uneven p a r a l l e l bars (George, 1980). In the introspection study, congruence between the group's introspections and t h e i r eye movements was determined. No attempt was made to specify d i r e c t i o n a l differences between the groups except that they were expected to d i f f e r i n t h e i r i n t r o s p e c t i o n about the body segments they f e l t contributed most to a successful performance. As i s evident, a l l of the three studies r e l y on the gymnast's v i s u a l perception as a way of gaining entry to t h e i r knowledge structure of the s k i l l s they perform. The research framework i s unique i n that i t suggests that an athlete's v i s u a l perception of •the events within t h e i r sport provides i n s i g h t to t h e i r own performance. Consequently, i t i s suggested that as s k i l l develops the gymnast stores an abstract i n t e r n a l representation that guides t h e i r perception of the event. It w i l l be shown, both t h e o r e t i c a l l y and e m p i r i c a l l y , that an athlete's perception may be viewed as a "mind's eye" of t h e i r actions . D. V i s u a l perception as the "mind's eye" of action The notion that the v i s u a l and motor systems share aspects of a common i n t e r n a l representation i s described by Turvey (1973, 11. 1977b) as a "constancy function." The intimacy of perceiving and acting i s i l l u s t r a t e d by comparing our p h y s i c a l a b i l i t y to write an 'a' with our a b i l i t i e s to perceive one. We are able to discern an 'a' as an 'A' even i f i t appears i n a thousand d i f f e r e n t forms. We are also able to write a myriad acceptable 'a' forms, none of which are exactly a l i k e , but nonetheless are perceived i n a constant manner. As w e l l , our a b i l i t y to write an 'a' i s not r e s t r i c t e d to a s p e c i f i c group of muscles. Indeed, we can 'write' i f c a l l e d upon with e i t h e r hand, our foot, even our nose. Obviously, some type of c e n t r a l l y stored representation must e x i s t that guides our movements i r r e s p e c t i v e of the muscle groups involved. We appear to have a c e n t r a l l y stored representation f o r the l e t t e r form that i s both perceived and produced i r r e s p e c t i v e of f e a t u r a l d i v e r s i t y and function. Of l a t e , the number of t h e o r e t i c a l and empirical bases supporting the idea that perception and action are c l o s e l y linked has grown dramatically (Gibson, 1979; Lee, 1978; Johannson, 1979, 1983; Neisser, 1976, 1983; Turvey, 1977a, 1977b; Reed, 1982; Haber, 1983; Newell and Barclay, 1982). Although these researchers vary i n the s p e c i f i c o r i e n t a t i o n they use, a c e n t r a l notion put f o r t h by a l l i s that v i s u a l perception develops i n concert with the developing motor system. As a consequence of perceiving and acting within defined environments, a type of knowledge i s developed that has two important components—as we become attuned to what i s "out 12. there" i n our world (exteroceptive information), we are also perceptually s e n s i t i v e to our bodies within that world (proprioception or k i n e s t h e s i s ) . These two sources of information do not e x i s t independent of one another. Rather, they are integrated to form a l i t t l e - e x p l o r e d type of information that i s c e n t r a l l y represented and c a l l e d here, knowledge structures of a c t i o n . In exploring the nature of t h i s type of movement information, there i s l i t t l e concern for d e f i n i n g the s p e c i f i c receptor or receptors that are stimulated i n the course of perception. I t i s not considered necessary to i s o l a t e the response or behaviourial mechanisms. Instead, the focus of the research concentrates on defining the nature of the information that i s perceptually s a l i e n t or meaningful to the subject. In t h i s research, therefore, the intent i s to detect and describe differences between the groups of gymnasts i n the way they handle information about gymnastics. Perceving pictures of human movement. Two studies by Sekiyama (1982, 1983) have shown that observers are perceptually s e n s i t i v e to pictures portraying d i f f e r e n t o r i e n t a t i o n s of the human hand. In the f i r s t study, subjects were shown pictures of the human hand viewed from d i f f e r e n t angles, one at a time. The hand presented was e i t h e r a r i g h t or l e f t hand and appeared i n eight d i f f e r e n t or i e n t a t i o n s which varied by angle of r o t a t i o n at the w r i s t . The subjects were asked to i d e n t i f y the hand as being e i t h e r r i g h t or l e f t . The subjects were prevented from moving t h e i r hands which were covered. Sekiyama reasoned that i f the movement of the hand were i n t e r n a l l y represented, the subjects would mentally move t h e i r hands into the p o s i t i o n s shown and i t would take longer f or them to respond to pictures where the hand was rotated through a greater number of degrees or i f i t would be d i f f i c u l t to move one's own hand into the portrayed p o s i t i o n . As expected, reaction time varied systematically according to the angular of departure from the upright. As w e l l , the reaction time function for the l e f t and ri g h t hand were not i d e n t i c a l but were mirror reversed from one another. That i s , the l e f t hand could be mentally rotated clockwise more e a s i l y than counterclockwise, and the r i g h t hand could be rotated counterclockwise more e a s i l y than clockwise. This was termed the "manageable d i r e c t i o n " of movement i n each hand. Sekiyama (1982, p.91) concluded that the subject's judgments were "based on a mental analogue of the actual movements of t h e i r own hands. Such a mental analogue would preserve k i n e s t h e t i c or proprioceptive information attending the movements." In the second study, Sekiyama (1983) asked subjects to p h y s i c a l l y model the shown hand posit i o n s and rate the degree of d i f f i c u l t y i n achieving each p o s i t i o n . The measured d i f f i c u l t y was then compared to the reaction time function obtained i n the f i r s t study. The ratings for degree of d i f f i c u l t y of the actual hand movements were s i m i l a r to the reaction time functions obtained i n the f i r s t study. This established that the r e s u l t s i n the f i r s t study were due to i n t e r n a l l y represented k i n e s t h e t i c factors rather than v i s u a l f a m i l i a r i t y with the stimulus information. Freyd (1983) has shown that a photographic presentation of someone undergoing a u n i d i r e c t i o n a l change i n motion, cues within an observer, an i n t e r n a l representation for the movement that i s perceived as continuing i n the d i r e c t i o n portrayed. She chose as stimulus material i r r e v e r s i b l e a ction sequences, for example, sequential photographs of someone jumping from a w a l l . Subjects were f i r s t shown a photograph of the i n d i v i d u a l i n m i d - f l i g h t followed by a second photograph i n one or two p o s i t i o n s — j u s t before the f i r s t p i c t u r e or j u s t a f t e r i t . A l l were asked to make judgments whether the second photographs occurred before or a f t e r the f i r s t . Freyd reasoned that i t would be more d i f f i c u l t f o r subjects to discriminate between the photographs taken s l i g h t l y a f t e r since the subjects had "unfrozen" the motion i m p l i c i t i n the i n i t i a l photograph and had already anticipated forward motion. Backward di s c r i m i n a t i o n would be easier as the subjects would not be expected to generate a type of information that i s e s s e n t i a l l y unique and not represented, i . e . , people do not perform the portrayed actions i n reverse. As expected, i t was harder to discriminate between the photos that were i n the " r e a l world temporal order." This i s an example of a s t a t i c stimulus 15. portraying human action inducing a dynamic representation, the arousal of k i n e s t h e t i c information. Freyd concluded that "people represent motion when viewing s t a t i c s t i m u l i . This, i n turn, supports the notion that perception and representation of motion play a c e n t r a l organizing r o l e i n the mind" (p.580). The Sekiyama and Freyd studies i l l u s t r a t e that the i n t e r n a l process of judging the p o s i t i o n of the body i s analogous to the external process of the actual movement. Both the i n t e r n a l process and the ph y s i c a l movements are accompanied by proprioceptive or ki n e s t h e t i c information and r e s t r i c t e d by i t . In Sekiyama's studies, the subjects mentally moved t h e i r " i n t e r n a l hand" i n the "manageable d i r e c t i o n . " In the Freyd study, the subjects " i n t e r n a l l y moved" i n the ant i c i p a t e d and only probable d i r e c t i o n . In the present study, i t i s anticipated that the gymnastic observers w i l l " i n t e r n a l l y move" as they view each sequence. Because they possess three d i s t i n c t s k i l l l e v e l s , i t i s anticipated that the k i n e s t h e t i c information they have i n t e r n a l l y represented w i l l d i f f e r as a function of t h e i r a b i l i t y l e v e l i n gymnastics. Therefore, i t i s expected that the groups w i l l focus t h e i r a ttention to d i f f e r e n t body segments i n the eye movement study, w i l l record d i f f e r e n t time and error scores i n the resequencing study and introspect d i f f e r e n t l y about the portrayed sequences. 16. Perceptual^invariants: Exploring information i n dynamic events The science of how observers perceive human movement i s c a l l e d "event perception." "Event perception" i s defined by Johansson (1979, p.94) as "the perception of any change of q u a l i t y , quantity, or p o s i t i o n during a chosen time i n t e r v a l . " There are i d e n t i f i a b l e properties about events of the human i n action that a f f o r d i t s recognition or i d e n t i f i c a t i o n . These properties are c a l l e d "perceptual i n v a r i a n t s . " One of the fundamental ideas advanced i n "event perception" i s that there are c e r t a i n advanced properties of the input a v a i l a b l e to the v i s u a l system that are inherent and are termed " i n v a r i a n t . " These properties are n a t u r a l l y perceived by what Runeson (1977) has termed "smart" perceptual systems. The nature of the difference between the groups of gymnasts i n viewing the sequences i s revealed through the detection and study of perceptual i n v a r i a n t s of dynamic events. Our focus i s to i d e n t i f y perceptual i n v a r i a n t s that are s p e c i f i c to three s k i l l l e v e l s i n gymnastics. In v i s u a l event perception there i s a constant flow of information over a receptive f i e l d of the r e t i n a followed by a continuous neural response. Of i n t e r e s t i s the e f f e c t of a change i n information over the r e t i n a . To i l l u s t r a t e , when a single dot i s moved i n a c i r c u l a r pattern on a video monitor, one's perception of the moving dot w i l l vary according to the cycle time of the dot. 17. If the cycle time i s short, one does not perceive a s e r i e s of dots but instead a continuous c i r c l e appears to be i n s c r i b e d upon the the screen. Perceptual mechanisms do not successively add each p o s i t i o n of the dot but instead perception of the event i s h o l i s t i c and s e l f evident, i . e . , "we experience conversation of s p a t i a l information over time as a single perceptual act and that perception i s i n f a c t "remembering" as well as "recording" (Johannson, 1979, p.98). S i m i l a r l y , as the gymnastic groups view the slow motion s l i d e presentation of the sequences, they do not see each s l i d e as a separate and d i s t i n c t e n t i t y . Instead, the movements are perceived as a whole, as a s i n g l e perceptual event. When viewing a dynamic event, there are s p e c i f i c and unique invariants that a f f o r d i t s recognition. V e r i f i c a t i o n of t h i s p r i n c i p l e was provided i n a study by Johansson (1975) i n which actors were f i t t e d with 12 small l i g h t s situated at the p r i n c i p a l j o i n t s . When they remained s t i l l i n a darkened room, naive observers (school children) were mystified by the presence of a random meaningless array or " i c o n " of l i g h t s . In a l l cases, the instant the actor moved, however, recognition was instantaneous and accurate. And i t did not matter whether the actor was human or animal, was walking, jogging, moving toward or away or dancing, the subject's perception was quick and accurate. Indeed, the more complex the event, the more e f f e c t i v e the sensory decoding. The detection of invariants i n dynamic events, as shown i n 18. Johansson's experiments, has been elaborated upon to i d e n t i f y i n v a r i a n t s i n gai t perception. There appears to be something perceptually unique about the gait of those we know that permits instantaneous i d e n t i f i c a t i o n even i n a crowd, or at a distance, or i n impoverished conditions such as point l i g h t d i s p l a y s . Cutting and Kozlowski (1977) created dynamic point l i g h t displays of friends walking. Through t h i s technique they were able to remove a l l non-gait sources of information such as f a m i l i a r i t y and shape cues leaving j u s t an undulating or walking array of l i g h t s . In a pre-post s i t u a t i o n , the subjects who knew one another returned two months l a t e r and were asked to i d e n t i f y who was i n each tape. T h i r t y eight percent of the tapes were c o r r e c t l y i d e n t i f i e d (chance was 16.7%). Using the same technique, Cutting and h i s colleagues have also established that observers of point l i g h t displays can accurately i d e n t i f y the gender of a subject. With only two seconds of viewing time, or four complete steps, i d e n t i f i c a t i o n was correct 67% of the time. Cutting's explanation f o r these findings has been the i d e n t i f i c a t i o n of a biomechanical invariant c a l l e d "center of moment." The center of moment, also c a l l e d the centre of movement, i s i d e n t i f i e d as "a fun c t i o n a l point within the topography of the object, around which o b j e c t - r e l a t i v e dynamics occur" (Cutting and P r o f f i t t , 1981). The center of moment, being concerned with d i s t r i b u t i o n of movements i s not the centre of gr a v i t y , which i s 19. concerned with the d i s t r i b u t i o n of mass, although the two may coincide. The center of moment i n walking i s an abstract l o c a t i o n within the body defined r e l a t i v e to movement of the shoulders and hips during the step c y c l e , i . e . , " i t i s a point within the walker around which everything moves" (Cutting and P r o f f i t t , 1981, p.259). Thus, hypothetical stress l i n e s drawn diagonally from shoulder to hip i n t e r s e c t at a point i d e n t i f y i n g that person's center of moment. Computer generated male and female synthetic walkers were developed i n the s a g i t t a l plane using hip, shoulder and torso measures of a large sample of walkers that captured the o s c i l l a t i n g a c tion of the shoulders and hips. Gender was c o r r e c t l y i d e n t i f i e d i n 80% of t r i a l s with center of moment accounting f o r 75% of the variance. The advantage of using synthesized walkers lay i n the a b i l i t y of the experimenter to generate s t i m u l i that d i f f e r e d only i n the l o c a t i o n of the center of moment. The center of moment appears to be a dynamic perceptual invariant that we use i n making judgements about the gender of i n d i v i d u a l s . Todd (1983) has i s o l a t e d the perceptual invariant that underlies our a b i l i t y to d i s t i n g u i s h walking from running. Computer simulations of g a i t were developed a f t e r the method developed by Cutting (1978). T h i r t y two d i f f e r e n t simulations were prepared by using a l l possible combinations of seven i d e n t i f i e d l e g functions for running and walking. That i s , a walking lower leg function was 20. combined with a running upper leg function, etc. Subjects were asked to i d e n t i f y whether the movement was running or walking and to also judge the q u a l i t y of the move on a r a t i n g scale, 1 to 5. The r e s u l t s were "overwhelming" i n showing that the invariant for the i d e n t i f i c a t i o n of running and walking was the lower leg function. And, Runeson and Frykolm (1981) have shown that i t i s a l l but impossible for an i n d i v i d u a l to pretend to l i f t a weight and not be detected. Observers watched point l i g h t displays of people l i f t i n g boxes of varying weights and were asked to guess how much weight was being l i f t e d . They then l i f t e d the boxes themselves estimating the weight. There was l i t t l e d i f f e r e n c e i n the two estimates i n d i c a t i n g that even impoverished perceptual input i s as accurate at estimating mass as performing the act personally. F i n a l l y , an unpublished study by Warren, reported by Mace (1983) shows that a person's perception of an event i s to a degree rel a t e d to t h e i r own dynamics. Groups of t a l l (193 cm., 6'2") and short (162.5 cm., 5'6") people were shown geometrically defined s t a i r s t e p s with r i s e r s of d i f f e r i n g heights. They were f i r s t asked to make a judgement about when i t would be impossible to climb the s t a i r s with a normal stepping motion and, secondly, to judge the height of the r i s e r they f e l t was optimal for performing work. Warren reasoned that a person's judgement i n both instances would be re l a t e d to t h e i r own leg length. He calculated a biomechanical maximum f o r the r i s e r height that took into account the leg length of short and t a l l subjects. This r a t i o , c r i t i c a l r i s e r height/leg length, turned out to be a constant .89. In judging the l i m i t of c l i m b a b i l i t y of the s t a i r s both the t a l l and short groups selected s t a i r s that, when re l a t e d to t h e i r leg height, approximated the .89 value of the predetermined r a t i o . In the second part of the experiment, determining the "optimal r i s e r height," Warren f i r s t determined the optimal r i s e r height for t a l l and short people by c a l c u l a t i n g t h e i r oxygen consumption while performing work on a s t a i r c a s e t r e a d m i l l with v a r i a b l e r i s e r . The optimal r i s e r height was determined as a r a t i o of r i s e r height/leg length and was .257 f o r t a l l people and .261 f o r short people. In making perceptual judgements about "optimal s t a i r s " the t a l l and short groups were very close to the optimal value of .26, "the best s t a i r s seem to be those that are approximately a quarter of ones' leg l e n gth—and people 'know' t h i s f a c t " (Mace, 1983, p.154). These studies i l l u s t r a t e that observers of human action are s e n s i t i v e to and make accurate decisions about the i d e n t i t y of people, t h e i r gender, and the types of movements they perform based upon perceptual i n v a r i a n t s . It i s suggested that as s k i l l i s developed i n a sport such as gymnastics, perceptual inva r i a n t s also evolve. What the e l i t e , intermediate and novice gymnast i s perceptually attuned to w i l l vary, t h i s variance being a function of t h e i r developing muscle system and the type of knowledge structure they have developed f o r each s k i l l . I I . Method A. Eye movement measures: a method for detecting differences i n  knowledge structures of action Eye movement measures are described as "a remarkably convenient methodological t o o l i n the sense that the l o c a t i o n of gaze constitutes an overt, e c o l o g i c a l l y v a l i d measure of the focus of attention at any given i n s t a n t " (Loftus, 1982, p.259). Location of gaze or a f i x a t i o n occurs when the subject's eye i s stationary with the fovea pointed at a s i n g l e part of the scene. F i x a t i o n duration or the length of gaze v a r i e s on average from 200 to 500 msec, with 1 to 5 degrees of information being processed. Saccades are the quick movements that separate f i x a t i o n s . They are the jumps of the eye from one f i x a t i o n point to another with 3 saccades per second characterizing normal scanning. A combination of f i x a t i o n s and saccades r e s u l t s i n the subject's scan path or the c h a r a c t e r i s t i c way the i n d i v i d u a l takes i n information from the scene presented. Yarbus (1967) has shown that a large proportion of f i x a t i o n s are d i r e c t e d to areas that were previously i d e n t i f i e d as highly informative. For example, people tended to f i x a t e faces and within faces the eyes and mouth. Fixations are not n e c e s s a r i l y drawn to areas of greatest d e t a i l or to contours, i . e . , the boundaries of 23. perceptual f i g u r e s . Scan patterns are also influenced by the type of d i r e c t i o n s given to subjects or t h e i r purpose i n viewing the scenes. Two conditions are most common. Subjects are asked to "free view" the s t i m u l i followed by r e c a l l or recognition tasks a f t e r viewing, or subjects are asked to watch f o r or attend to s p e c i f i c elements. A free viewing protocol was used i n the present study. Noton and Stark (1971), using a free viewing condition, found that subjects develop c h a r a c t e r i s t i c scan patterns when viewing l i n e drawings of faces or objects, that once established during a learning condition, reappeared i n subsequent viewing of the same material. They c a l l e d t h i s a "feature r i n g a sequence of sensory and motor memory traces, a l t e r n a t e l y recording a feature of the object and the eye movements required to reach the next feature" (p.39). They r e l a t e d feature rings to the subject's i n t e r n a l representation f o r that object. During recognition the i n t e r n a l representation i s "matched s e r i a l l y with the object, feature by feature, e i t h e r with eye movements or with i n t e r n a l s h i f t s of attention; the attention s h i f t s connect the features i n a scan path, which i s usually followed when v e r i f y i n g the features during recognition" (p.42). Stark and E l l i s (1981) have shown that "eye movements are co n t r o l l e d by cognitive models (either i d e a l s or experimentally generated) already present i n the b r a i n " (p.193). Subjects viewed 24. types of s t i m u l i that had a l t e r n a t i v e perceptual i n t e r p r e t a t i o n s , for example, a t r i p l e ambiguous fig u r e i n which an old man, an old woman and a young g i r l are simultaneously portrayed. In viewing the ambiguous f i g u r e , the subjects were asked to s i g n a l when they "saw" each of the faces while t h e i r eye movements were recorded. Three d i s t i n c t scan patterns were observed corresponding to each of the faces. As w e l l , i n reviewing the f i g u r e on l a t e r occasions, the same types of scan patterns for the faces were observed for the same subject. In the present study, the type of stimulus materials used was a dynamic s e r i e s of s l i d e s each varying only s l i g h t l y from the previous one, i . e . , the slow motion (3 sec. per s l i d e ) presentation of a moving body. Stark and E l l i s (1981) have shown that, across a s e r i e s of sequential s t i m u l i , subjects e s t a b l i s h scan patterns i n which p a r t i c u l a r items of information are f i x a t e d more often than others. A p i l o t ' s scan pattern was assessed as he observed a changing display of a i r t r a f f i c information. The o v e r a l l display included the p i l o t ' s own ship, an intruder and ground referenced symbols that were updated to display changing encounters. The p i l o t ' s d i r e c t i o n of gaze was recorded over 24 updates each observed for 28 seconds. They found that despite the changing or dynamic nature of the display, the p i l o t i d e n t i f i e d an area that was perceptually more s a l i e n t than a l l others. They c a l l e d t h i s the "home base" phenomenon. 25. S i m i l a r l y , we may ask whether across a series of s l i d e s d i s p l a y i n g gymnastic information, are there body segments that draw the focus of attention of the e l i t e , intermediate and novice gymnasts groups i n d i f f e r i n g ways? If t h i s occurs, then i t i s s t i l l necessary to determine what t h i s i s r e f l e c t i v e of. There are currently two i n t e r p r e t a t i o n s possible as provided by eye movements research. F i r s t , the gymnast may be drawn to observe a part of the movement because i t i s unusual or unexpected, the observed performance i s a t y p i c a l , not well done or out of context. A l t e r n a t i v e l y , the gymnast i s drawn to the usual or expected. Their f i x a t i o n s are r e f l e c t i v e of t h e i r i n t e r n a l awareness of the knowledge structure they possess f o r that movement. Mackworth and Morandi (1967) and Loftus and Mackworth (1978) supply evidence for the f i r s t p o s i t i o n . They reported that observers are drawn to observe the unpredictable or the unusual. In a r r i v i n g at t h i s conclusion, however, they used v i s u a l s t i m u l i that presented l i t t l e coherent meaning to the subjects, f o r example, a p i c t u r e of a c o a s t l i n e segmented into squares much l i k e a jigsaw puzzle or scenes i n which were placed u n l i k e l y objects, for example, an octopus i n a farm yard. In these studies there i s no question that i f unusual or contextually inappropriate information i s supplied to a subject, they w i l l focus t h e i r a t t e n t i o n to the a t y p i c a l areas. But i f v i s u a l materials are selected that are normal and usual, the eye movement data obtained 26. is reflective of the subject's knowledge and understanding for what is portrayed. Antes (1974) illustrated this using meaningful pictures from the Thematic Apperception Test that were previously rated for units of high and low informational content. He found that high information areas were fixated early, but for a brief period of time followed by low information areas that were fixated relatively longer. He concluded that subjects f i r s t identify what for them is the most relevant piece of information in the scene, they then direct their gaze in search of supporting or confirming information. Therefore, in scenes where the unusual is present, subjects are immediately sensitive to this and direct their attention accordingly. In situations where a concerted attempt has been made o to present normal or typical information that is contextually appropriate, as i s the case with the presentation of the gymnastic material selected here, subjects are drawn to confirm expectations, expectations that are intimately tied to their internal representation or knowledge structure for that material. Eye movements have been used to characterize differences in perceptual scan patterns between expert (or experienced) and novice (or inexperienced) athletes. Bard and Fleury (1976) showed varsity and novice basketball players slides from basketball games depicting strategic maneuvers. Eye movement measures showed the varsity basketball players focussed their attention upon the opponent or on open spaces with the novices tending to observe t h e i r own partner. Bard, Fleury, C a r r i e r e , and Halle (1980) found that expert and novice gymnastic judges d i f f e r e d i n the way they observed the performance of gymnasts. The expert judges observed the performers' upper body while the novice judges focussed t h e i r a ttention to the legs. Also, i n both the basketball and the gymnastic study, the novices recorded more f i x a t i o n s than did the experts. B. The resequencing task: detecting differences i n knowledge  structures of act i o n The resequencing task i s a problem solving task that required the subjects to restore the temporal order of a gymnastic sequence that had been randomly misordered. I t was hypothesized that against constraints of time and accuracy, the more highly s k i l l e d gymnasts w i l l function i n a conceptually driven or top-down manner i n contrast to the more data driven or bottom-up processes of the intermediates and novices. Conceptually driven and data driven perceptual systems. In the resequencing task there i s a large number of int e r p r e t a t i o n s or competing organizations possible with only one being correct. When the objective of the task i s thus made ambiguous, the v i s u a l 28. resequencing task t e s t s whether the gymnast draws not only upon the surface information or data that i s equally a v a i l a b l e to a l l i n the photographs but also accesses an i n t e r n a l structure that has developed as a consequence of s k i l l attainment. Conceptually driven systems are guided by i n t e r n a l representations or knowledge structures for a p a r t i c u l a r domain. Data driven systems process the information at hand and attempt to make sense of v i s u a l material based upon observable surface features. When i n d i v i d u a l s are l a r g e l y data driven they process the information presented before them without any pre-conceived idea about what i s most important. Consequently, they have a reduced a b i l i t y to discriminate or make judgements r e l a t i v e to the task at hand. In contrast, when i n d i v i d u a l s are guided by knowledge structures "the subject s t a r t s with a conceptualization of what might be present and then looks for confirming evidence, b i a s i n g the processing mechanisms to give the expected r e s u l t " (Lindsay and Norman, 1977, p.13). Conceptually driven systems are recognized as being more e f f i c i e n t and more accurate i n processing information, e s p e c i a l l y i f that information i s complex, with abstract or t a c i t meaning. C. The i n t r o s p e c t i o n task The i n t r o s p e c t i o n task was designed to explore the extent of the r e l a t i o n s h i p between a gymnast's int r o s p e c t i o n about a s k i l l 29. and t h e i r actual eye movements r e l a t i v e to that s k i l l . Introspection i s defined as the objective d e s c r i p t i o n of conscious content i n terms of i t s elements and a t t r i b u t e s or a l t e r n a t i v e l y a d e s c r i p t i o n of one's experiences or patterns of behavior. Although there has been l i t t l e a t t e n t i o n paid to an athlete's verbal assessment of t h e i r own motor s k i l l s , when bolstered by eye movement data and other tasks, i t may prove to be an e f f e c t i v e v e h i c l e f o r gaining i n s i g h t into an athlete's knowledge about acti o n . D. Subjects and design T h i r t y subjects were selected and assigned to three groups based on t h e i r expertise i n gymnastics. Groups one and two, the e l i t e and intermediate gymnasts, were selected from a n a t i o n a l l y ranked gymnastic club. Group three, the novice gymnasts, were members of a soccer team. A l l subjects were female. The e l i t e and intermediate gymnasts were assigned to groups as ranked by t h e i r club. The members of the e l i t e group were a l l n a t i o n a l l y ranked a t h l e t e s , with one subject being ranked number one i n Canada. The intermediate gymnasts were a l l r e g i o n a l l y ranked competitors. The t h i r d group were accomplished soccer players but with l i m i t e d experience i n Olympic l e v e l gymnastics. Some members of the group, however, had been members of r e c r e a t i o n a l gymnastic clubs and one was on her high school team. 30. None of the novice group had experienced the i n t e n s i t y of t r a i n i n g and competition that the e l i t e and intermediate gymnasts had undergone. The e l i t e gymnasts were on average 13.2 years of age with 5.8 years i n gymnastic competitions. The intermediate gymnasts were 12.4 years of age and had been competing for 3.5 years. The mean age of the novice group was 13.6 years. They had 0.30 years experience i n gymnastic competition but had 5.8 years i n soccer competition. The e l i t e and intermediate gymnasts had competed i n soccer for an average of 0.95 years. E. The eye movement study Apparatus. The Gulf and Western Applied Sciences Laboratory Eye View Monitor (Model 1994) was used to observe the eye movements of the subjects. The v i s u a l s t i m u l i were shown by a standard Kodak ektachrome s l i d e projector and were rear projected on a screen lm. x lm. i n s i z e positioned 1.5 metres i n front of each subject. The set up resulted i n a viewing f i e l d of 84 degrees, f r o n t a l . The monitor was f i r s t c a l i b r a t e d f o r each subject's head and eye movements i n the v i s u a l f i e l d by using a c a l i b r a t i o n s l i d e composed of a 3 x 3 matrix of nine numbers. The Gulf and Western system automatically corrects for 16.39 cubic mm. of head motion thus freeing the subjects from unusual head constraints and also greatly reducing the set-up time. A chin rest was used as i t added to the comfort and s t a b i l i t y of each subject. An i n v i s i b l e collimated beam of i n f r a - r e d l i g h t was r e f l e c t e d from the eye providing a corneal r e f l e c t i o n that also illuminated the p u p i l by r e f l e c t i n g l i g h t from the back of the eye. The corneal r e f l e c t i o n , or f i r s t Purkinje image, i s the h i g h l i g h t i n the eye that i s r e a d i l y observable by looking into the eye of another. This image was picked up by an i n f r a - r e d s e n s i t i v e t e l e v i s i o n camera as a bright p u p i l with an even br i g h t e r corneal r e f l e x . A computer monitoring the video scan s i g n a l joined the centroids of the p u p i l and the corneal r e f l e x into a vector. This displacement vector, or directed l i n e of magnitude, provided the observer's l i n e of sight r e l a t i v e to the scene. The structure of the eye and the nature of the corneal r e f l e x permits computation of the vector. The corneal r e f l e x appears always i n l i n e with the centre of corneal curvature. When the head moves, the corneal r e f l e x and the p u p i l centre move within l i m i t s together. I f the head i s s t a b i l i z e d , as i n the s i t u a t i o n here, the corneal r e f l e c t i o n moves d i f f e r e n t i a l l y with respect to the p u p i l centre and systems such as the Gulf and Western automatically f i n d the p u p i l centre and corneal h i g h l i g h t as the eye rotates to look at p o s i t i o n s i n the v i s u a l f i e l d (Young and Sheena, 1975; Sheena and Flagg, 1978; Cummings, 1978; Hainline, 1981). 32. Three cameras were used. A face camera permitted the correct p o s i t i o n i n g of the eye. An i n f r a - r e d eye camera formed an image of the corneal r e f l e c t i o n and p u p i l . The scene camera photographed the gymnastic information. The outputs of the eye camera and the scene camera were combined producing data as shown i n Figures 2a and 2b. Eye f i x a t i o n s appear as x/y coordinates or cross-hairs superimposed on a video tape of the scene being scanned at a rate of 60 samples per second. Insert Figure 2 about here _--^&-A*5&^-_i\\£_ S t i m u l i . Six sets of s l i d e s were prepared portraying world cl a s s gymnasts performing compulsory Olympic movements. Figure 3 shows 12 of the 26 s l i d e s that comprised the beam handstand sequence. Insert Figure 3 about here _ _ ^ ? r _ A r 5 f ^ i _ A V 3 _ The sets of s l i d e s were developed using a microphotography technique that extracted the maximum action component of each frame of a 16 mm. f i l m . The gymnastic f i l m s e r i e s selected was produced by the A t h l e t i c I n s t i t u t e (1982). These fi l m s were chosen over a wide s e l e c t i o n of others as they had been produced recently and the performance of the athletes was t e c h n i c a l l y of a high q u a l i t y . The fil m s were shot at a f a s t speed r e s u l t i n g i n very l i t t l e b l u r r i n g 33. of the action. They were taken from e f f e c t i v e v i s u a l angles r e s u l t i n g i n a large body image. The films were i n color, and the f i l m s e r i e s was complete, featuring a l l events i n women's gymnastics. A l l of these factors contributed to the development of s t i m u l i that were t e c h n i c a l l y of a high q u a l i t y . The s l i d e s were selected on the basis of two c r i t e r i a . F i r s t , sequences of low to medium d i f f i c u l t y were chosen with a l l of the e l i t e and the majority of the intermediate gymnasts being able to perform the movements ro u t i n e l y . Very few of the novices could perform these s k i l l s . Secondly, i n s e l e c t i n g the frames, the time course of the movement was preserved by s e l e c t i n g a constant 1 i n 5 or 1 i n 3 frames with t h i s r a t i o varying according to the speed of the movement. Since some of the movements were performed much fa s t e r than others (e.g., handspring vs. a walkover move on the beam), a t i g h t e r r a t i o of frames s e l e c t i o n was necessary. The name of each sequence with number of s l i d e s was as follows: f l o o r exercise - front walkover (22), f l o o r exercise -forward handspring (13), uneven p a r a l l e l bars - g l i d e k i p , stoop through (18), beam - back walkover through stag p o s i t i o n (22), beam - cartwheel to handstand, quarter turn to handstand s p l i t (26), and va u l t - h a l f on, h a l f o f f (17). Each sequence was introduced by an introductory s l i d e that f i r s t named each movement. The order of presentation of the s i x sequences was as above, front walkover f i r s t and v a u l t l a s t . T i t l e or introductory s l i d e s were used to 34. decrease the recognition component of the task providing the subjects with 9 sec. to r e c a l l each movement. Procedure. Procedural i n s t r u c t i o n s to the subjects were neutral and non d i r e c t i o n a l — a free scanning s i t u a t i o n as follows: You are going to be shown a number of s l i d e s showing gymnastic movements. We would l i k e you to watch each sequence of s l i d e s very c a r e f u l l y as you w i l l be asked to complete a number of tasks when f i n i s h e d . Viewing time per s l i d e was 3 sec. with an i n t e r s l i d e i n t e r v a l of 0.5 sec. The introductory/recognition phase of each sequence was 9 sec. The t o t a l viewing time f o r a l l s i x sequences was 9.5 minutes. A l l eye movement data were c o l l e c t e d by the same two inves t i g a t o r s who maintained consistent ro l e s throughout the study. Immediately following the completion of the eye movement task, the resequencing and int r o s p e c t i o n tasks were given. T o t a l t e s t i n g time f o r each subject averaged 1.5 hours. Four consecutive days were required to complete the data c o l l e c t i o n process. F. The resequencing study Six sets of photographs were derived from the s i x sets of s l i d e s used i n the eye movement study. Each sequence was reduced to 11 or 12 photos, 10 cm x 7.5 cm i n s i z e . I t was necessary to 35. shorten the sequences as they not only varied greatly i n the number of s l i d e s per sequence but also had proven to be too unwieldly i n p i l o t studies. Viewing the whole set of photographs rather than the shortened ser i e s resulted i n the subject's having to a c t i v e l y scan a very wide v i s u a l f i e l d , thus imposing a memory factor which was considered detrimental to the task. Procedure Three p i l o t studies resulted i n the use of the following test procedures. The order of presentation of each sequence was randomly determined by drawing numbers 1 to 6. The photographs were manually placed out of order on a table with the subjects seated on a swivel chair, t h e i r backs to the experimenter. Upon request, the subjects turned to face each sequence, t h e i r eyes closed. When they opened t h e i r eyes a d i g i t a l timer was started. The timer was stopped when the subjects v e r b a l l y indicated they were f i n i s h e d reconstructing each sequence. Two measures were recorded: t o t a l resequencing time and the t o t a l number of e r r o r s . Time measures were recorded to a 1/10 s e c ; errors were determined by using a s e r i a l order r e c a l l procedure. In t h i s procedure, an error i s defined as any d i g i t , word or, i n t h i s case, photo that i s sequentially misplaced. An i n c o r r e c t response may be recorded as an error only once; a s t r i n g of photographs misplaced within a sequence i s recorded as one error . A l l subjects were f i r s t given a p r a c t i s e t r i a l and the 36. procedures thoroughly explained before the experimental t r i a l s began. G. The in t r o s p e c t i o n study Two p i l o t studies were performed, which led to the following procedures. Upon completion of resequencing each set of photos, the subjects were asked to introspect about the i n t a c t movement displayed i n i t s correct order before them. Instructions to each subject were as follows: Imagine that you are performing the movement shown. Where i n the move do you f e e l i t i s most important to focus or concentrate your thoughts/attention when you are performing i n competition (or i n the case of the novices where they think one would have to focus t h e i r a t t e n t i o n i f they were performing). Two count data were obtained. One, the i d e n t i f i e d photographs were recorded by t h e i r corresponding eye movement's s l i d e number. The subjects were free to i d e n t i f y as many photographs as they wished. Within the selected photograph(s) they were then asked to i d e n t i f y the body segments they thought contributed most to a successful performance. Their responses were recorded using the same coding system used i n the eye movements study. 37. H. Eye movement and introspection coding system Eighteen categories were developed to categorize the body, equipment, background and s p a t i a l elements found within the v i s u a l s t i m u l i as shown i n Table 1. Insert Table 1 about here „ _ 3 _ ^ _ _ W s S . _ 3 J P r i o r to the c o l l e c t i o n of the data, a l l of these elements were anticipated except the l a s t , number 18. Figure 2a shows a t y p i c a l piece of eye movement data, the x/y coordinates showing a subject f i x a t i n g on the shoulder region of the competitor, thus coded number 02. Figure 2b shows a code 18. In t h i s s i t u a t i o n the subject i s f i x a t i n g a s p a t i a l or offbody dimension of the performance. I t became apparent that subjects often f i x a t e d space but were not viewing background features. This space was i d e n t i f i e d as movement space, with the speculation that subjects were eit h e r judging the q u a l i t y of a p o s i t i o n or a n t i c i p a t i n g the movement of the gymnast into the next p o s i t i o n . In order to maintain a degree of o b j e c t i v i t y f o r these f i x a t i o n s , the body segment nearest to the f i x a t i o n was recorded. Nearness was defined by holding a 2 degree di s c ( a c t u a l l y a 25 cent piece) on the f i x a t i o n at arm's length. The body segment cut most by the coin was recorded. I f no contact with a body segment was p o s s i b l e , the f i x a t i o n was then coded as 18, movement space. 38. Insert Figures 2a and 2b about here ^ £ _ - £ w _ W _ & i r _ \ _ \ J g . The coding of the tapes was done by the invest i g a t o r and required a considerable amount of care to e s t a b l i s h accuracy and r e l i a b i l i t y across the t h i r t y subjects. When more than one i n d i v i d u a l i s involved i n coding, the r i s k of confounding the data i s increased (Sheena and Flagg, 1978). The f i x a t i o n point when each s l i d e came on was not coded, as i t was probably not under stimulus c o n t r o l . I t requires 150-300 msec, f o r a subject to over t l y respond to a v i s u a l s t i m u l i (Young and Sheena, 1975). The f i r s t four f i x a t i o n s were coded as per Table 1. In addition, the t o t a l number of times the subject f i x a t e d each s l i d e was recorded. A r e l i a b i l i t y check was established by recoding randomly i d e n t i f i e d s l i d e s and then comparing these r e s u l t s against those f i r s t obtained. When a 100% agreement was obtained on an unlimited number of t r i a l s , the coding procedure was considered error f r e e . An overview of the data c o l l e c t i o n and count procedures i s provided by Table 1. Each s l i d e i n each sequence provided f i v e pieces of d a t a — f o u r f i x a t i o n s coded by body segment category and the t o t a l number of f i x a t i o n s per s l i d e . For each subject a t o t a l of 478, i . e . , 472 + 6 pieces of eye movement data were c o l l e c t e d . 39. The number of codes used va r i e d f o r each sequence due to t h e i r nature. For example, the beam moves required a l l 18 codes since the action of the legs assumed both t r a i l and lead leg as well as legs together p o s i t i o n s . In contrast, the f l o o r moves did not display a legs together o r i e n t a t i o n n e c e s s i t a t i n g the use of 15 codes. An eye f i x a t i o n was said to have occurred when each x/y coordinate point s h i f t e d from one category region to another among the category regions of each s l i d e . A l l subjects viewed each s l i d e f o r three seconds and only the f i r s t four f i x a t i o n s were coded. In the event that a subject f i x a t e d a s l i d e l e s s than four times (a very rare occurrence) then t h i s number was recorded. The t o t a l number of derived category v a r i a b l e s was 60 for the walkover and f l o o r moves (15 body categories x 4 f i x a t i o n s ) , 56 for the unevens and vault (14 categories x 4 fixations)and 72 each for the beam moves (18 categories x 4 f i x a t i o n s ) f o r a t o t a l of 376 category v a r i a b l e s describing each subject's focus of attention (see Appendices A to F ) . Rationale f o r data combination Because of the volume of data, the number of categories was reduced by combining the o r i g i n a l 14, 15 or 18 body segment codes. Six body segment un i t s were defined i n l i g h t of biomechanical and t e c h n i c a l considerations i n gymnastics. The s i x body segments, achieved by the combinations shown i n Table 40. 1, were l a b e l l e d head, shoulders, arms, hips, legs and offbody (see Appendices G to L ) . In a d dition, the f i x a t i o n s were combined i n two ways, a) the f i r s t two and l a s t two f i x a t i o n s were combined into two separate categories and l a b e l l e d f i r s t glance and second glance, r e s p e c t i v e l y . The three groups observed mean f i x a t i o n s to the s i x body segments for f i r s t glance and second glance for each of the body segment categories are shown i n Appendices M to R; the means for a l l four f i x a t i o n s combined are shown i n Appendices S to U. Support f o r these combinations was provided by a ser i e s of exploratory univariate analyses performed with the f i x a t i o n s combined. I t would always be possible to return to the o r i g i n a l data structure i f a more in-depth analysis was necessary. As w i l l be shown i n the following analyses, the data combination steps were undertaken without los s of the c r i t i c a l aspects of the o v e r a l l r e s u l t s . 41. I I I . Results A. The eye movement study: T o t a l sequence and phase analysis The eye movement data were analysed at two l e v e l s . The f i r s t a n a l y s i s , c a l l e d a t o t a l sequence an a l y s i s , i d e n t i f i e d d ifferences between the three groups i n eye f i x a t i o n s to body segments across a l l s l i d e s of each sequence. The observed v a r i a b l e to be analysed was derived by summing each body segment category across the t o t a l number of s l i d e s . The second an a l y s i s , the phase an a l y s i s , segmented or pa r t i t i o n e d each sequence into temporal phases of a continuous motor sequence. The temporal segmentation of each sequence was somewhat a r b i t r a r y as i t i s not known how performers segment actions although there i s much speculation about how t h i s i s accomplished from both t h e o r e t i c a l and applied sources (Glencross, 1978; George, 1982). In a phase analysis the sequence was pa r t i t i o n e d into beginning, mid, and end or more temporal u n i t s . The objective i n the phase analysis was to determine whether the athletes showed consistent f i x a t i o n s to body segments across a l l phases of the sequence or to i d e n t i f y differences i n f i x a t i o n s that accrued as a r e s u l t of the d i f f e r e n t phases. 42. T o t a l sequence an a l y s i s : Summary The t o t a l sequence analysis examined, i n depth, the eye movement responses i n s i x sequences i n women's gymnastics with differences explained r e l a t i v e to s i x body segments (see Tables 9 and 10). S i g n i f i c a n t differences (p_(.05) i n eye f i x a t i o n s between e l i t e and the intermediate/novice groups of gymnasts were observed i n four of the six sequences. The e l i t e athletes attended le s s to the head i n a l l of the sequences, except the forward handspring and unevens, and more to the hips i n the two beam sequences; the arms were attended more i n the v a u l t . The e l i t e s observed offbody factors l e s s i n the unevens than d i d the two other groups. In comparing the intermediates to the novices, the intermediates attended l e s s to offbody information i n a l l the sequences, except the forward walkover and handspring. They were drawn to observe the head more than the novices i n the handspring and the unevens, the shoulders more i n the unevens, and the legs more i n beam backover. The hips and arms were observed i n a s i m i l a r manner by the two groups. It was concluded that i n the s i x sequences the e l i t e , intermediate and novice gymnasts d i f f e r e d i n the way they perceived the head, hips and offbody, whereas they did not seem to d i f f e r as much i n t h e i r perception of the arms, shoulders and legs. In two moves, the walkover and handspring, differences between the groups were minimal. Two reasons are offered: a) they were t e c h n i c a l l y 43. the l e a s t d i f f i c u l t of the s i x sequences, and b) some of the novices could perform these moves. The groups did not d i f f e r i n t h e i r pattern of f i x a t i o n s from the f i r s t to the second glance. Neither did they d i f f e r i n the t o t a l number of f i x a t i o n s per s l i d e . Only the f i r s t four f i x a t i o n s i n each s l i d e were coded. However, there were i n most s i t u a t i o n s more f i x a t i o n s to each s l i d e than four. It was expected that the e l i t e athletes would record fewer f i x a t i o n s , a r e s u l t reported i n e a r l i e r studies (Bard and Fleury, 1976; Bard, Fleury, C a r r i e r e , and H a l l e , 1980). The groups, however, did not d i f f e r s i g n i f i c a n t l y i n t h e i r frequency of f i x a t i o n s . Results i n d e t a i l : M u l t i v a r i a t e Analysis of Variance. A 3 x 2 x 6 f a c t o r i a l (expertise x f i x a t i o n x body segment) mu l t i v a r i a t e analysis of variance was performed with repeated measures on the l a s t two f actors on each of the s i x sequences. There were three l e v e l s of expertise ( e l i t e , intermediate, novice), two l e v e l s of f i x a t i o n ( f i r s t glance, second glance) and s i x l e v e l s of body segment (head, shoulders, arms, hips, legs, offbody). Linear contrast analyses were performed on the set of s i x body segment category scores, using two sets of orthogonal contrast c o e f f i c i e n t s on the sample design f a c t o r of expertise and the response design f a c t o r of f i x a t i o n . The mean contrast e f f e c t s r e s u l t i n g from making sets of l i n e a r contrasts of the l e v e l s of the 44. two f a c t o r s , expertise and f i x a t i o n , are presented i n Tables 2 to 7 for the s i x sequences. Insert Tables 2 to 7 about here S p e c i f i c a l l y , the contrast e f f e c t s were determined i n a 3 x 2 matrix: three contrasts for expertise (constant, E-IN, I-N) and two contrasts for f i x a t i o n ( a l l four f i x a t i o n s and f i r s t glance compared to second glance). These contrast e f f e c t s determined whether the groups varied i n t h e i r observation behaviors, and i f so was t h i s due to t h e i r l e v e l of expertise, to the type of body segment, to f i x a t i o n or some i n t e r a c t i o n of these three f a c t o r s . The e l i t e gymnasts with the highest l e v e l of p h y s i c a l s k i l l were expected to possess a knowledge structure for each of the sequences that was d i s t i n c t from that of the other two groups. This d i s t i n c t i v e n e s s would be indicated by the f i r s t contrast, E-IN. S i m i l a r l y , the second contrast, I-N, was set out to i d e n t i f y and describe differences between the intermediate and novice groups. Contrasts were also made to explore the e f f e c t of f i x a t i o n i n two ways. F i r s t , the e f f e c t of combining a l l four f i x a t i o n s was explored; second, the e f f e c t of combining the f i r s t two f i x a t i o n s ( f i r s t glance) was compared to combining the l a s t two f i x a t i o n s (second glance). 45. Fixations to body segments i r r e s p e c t i v e of l e v e l of expertise. Because there were no preconceived p r o f i l e s f o r the frequencies of eye f i x a t i o n s to the s i x body segments, f i x a t i o n s to body segments were described as observed. No p a r t i c u l a r s t a t i s t i c a l tests were performed, except f or those i n d i c a t i n g that the constant part of the f i x a t i o n scores by s i x body segment categories are greater than zero. M u l t i v a r i a t e and uni v a r i a t e tests f o r t h i s were a l l s i g n i f i c a n t as shown i n Table 8. Insert Table 8 about here The subjects, i r r e s p e c t i v e of t h e i r l e v e l of expertise, did not view the s i x body segments i n a s i m i l a r manner. Instead preferences were shown for one body segment over another. This can be seen by comparing the r e l a t i v e s i z e s of the univariate F-values corresponding to the s i x body segment categories. For example, the leg part of the performing body drew the greatest amount of attention f o r a l l the sequences except the v a u l t . Offbody factors were observed l e a s t f o r a l l sequences except the uneven bars. The e l i t e gymnasts compared to the intermediates/novices. Table 9 presents the multivariate and univariate F values f o r the i n t e r a c t i o n of expertise and body segment i r r e s p e c t i v e of f i x a t i o n . 46. Insert Table 9 about here The E-IN contrast c l e a r l y showed that the e l i t e gymnasts viewed the beamhand and vault sequences i n a s i g n i f i c a n t l y d i f f e r e n t manner than did the intermediates and novices, beamhand, M u l t i v a r i a t e F(6,22) = 2.85, £<.03; vault M u l t i v a r i a t e F(6,22) = 2.83, p_<.03. Univariate analyses of the body segments revealed these differences to be i n focus of attention to the head F(l,27) = 6.89; hip, F(l,27) = 10.54; and offbody, F(1.27) = 4.76 of the beamhand and to the head, F(l,27) = 6.00) and arms, F(l,27) = 4.80 of the v a u l t . The d i r e c t i o n a l i t y of these differences was determined by r e f e r r i n g to the observed contrast e f f e c t s as shown i n the corresponding Tables 6 and 7, r e s p e c t i v e l y . In the beamhand, the e l i t e competitors focussed t h e i r attention more to the hip region of the performer than did the intermediate/novices. In contrast, the intermediate/novices were drawn to observe the head and offbody regions, areas that held very l i t t l e i n t e r e s t f o r the e l i t e gymnasts. In the v a u l t , the e l i t e gymnasts were more att e n t i v e to the arms and focussed very l i t t l e of t h e i r a ttention to the head region, a tendency opposite to that of the intermediate/novices. Although the other sequences were not covered by the umbrella of a s i g n i f i c a n t M u l t i v a r i a t e F, a number of univariate analyses were s i g n i f i c a n t and were supportive of the r e s u l t s found above. 47. The head was viewed i n a s i m i l a r manner i n the walkover, F(l,27) = 7.95 and beamwalk, _F(1,27) = 8.02 as were the hips i n the beam walk, F(l,27) = 10.15. The I-N contrast showed that the intermediate gymnasts d i f f e r e d from the novices i n t h e i r perception of three of the s i x sequences: the unevens, M u l t i v a r i a t e F(6,22) = 6.28 £<.0001; the beamhand, M u l t i v a r i a t e F(6,22) = 2.74 £<.04; and the v a u l t , M u l t i v a r i a t e F(6,22) =3.98 p<0001. Univariate analyses of the body segments indicated that the intermediate gymnasts focussed t h e i r a ttention more so to the head, F(l,27) = 5.40 and shoulders, F(l,27) = 6.63 i n the unevens. In four of the s i x sequences, the novices were s i g n i f i c a n t l y more attentive to factors i n the surround or offbody: the unevens, F(l,27) = 33.38; beamwalk, F(l,27) = 6.74; beamhand, F(l,27) = 14.71; vault F(l,22) - 14.44. The intermediates d i f f e r e d from the novices i n t h e i r focus of at t e n t i o n to the head, F(l,27) = 4.24 i n the handspring, and to the legs F(l,27) = 6.22 i n the beamwalk. In both instances, the intermediates were more attentive to these body segments than were the novices. Table 10 presents an o v e r a l l summary of the observed differences (p_^.15) i n f i x a t i o n to body segments i n the s i x sequences. Insert Table 10 about here 48. E f f e c t o f t h e f i r s t v s . t h e second g l a n c e . The s i x body segment c a t e g o r y s c o r e s were compared w i t h r e s p e c t t o t h e f i r s t and second g l a n c e o f eye f i x a t i o n s . T a b l e 11 p r e s e n t s t h e d i f f e r e n c e s , f i r s t g l a n c e v s . second g l a n c e , i r r e s p e c t i v e o f e x p e r t i s e . I n s e r t T a b l e 11 about h e r e __JD_ J^^ WJL_).S>1 The M u l t i v a r i a t e F v a l u e s were s i g n i f i c a n t f o r a l l sequences e x c e p t t h e unevens. U n i v a r i a t e a n a l y s e s o f body segments i s o l a t e d t h e s e d i f f e r e n c e s as shown. T h i s d i f f e r e n c e i s t o be e x p e c t e d due t o t h e i n h e r e n t n a t u r e o f v i s u a l s c a n n i n g w h i c h i n v o l v e d t h e s u b j e c t s s h i f t i n g t h e i r gaze f r o m one body segment t o a n o t h e r . F i r s t v s . second g l a n c e by body segment i n t e r a c t i o n The E-IN c o n t r a s t o f T a b l e 12 shows t h e r e were no s i g n i f i c a n t M u l t i v a r i a t e F v a l u e s f o r t h i s c o n t r a s t . I n s e r t T a b l e 12 about h e r e !^^^J^L_.>J-L3 The e l i t e group d i d n o t d i f f e r f r o m t h e i n t e r m e d i a t e / n o v i c e s i n t h e way t h e y o b s e r v e d t h e sequences f r o m f i r s t g l a n c e t o second g l a n c e . A s i m i l a r r e s u l t was e v i d e n t f o r t h e i n t e r m e d i a t e and n o v i c e I-N c o n t r a s t s . U n i v a r i a t e a n a l y s i s f o r each body segment d i d d e t e c t d i f f e r e n c e s f o r c e r t a i n o f t h e sequences. The e l i t e s d i f f e r e d f r o m 49. the intermediate and novices in the way they fixated the hip in the two beam moves and in the vault. The intermediates differed from the novices in fixating the head, arms, and hip of the walkover and handspring. Total number of eye fixations. The total number of fixations was recorded for each 3 sec. presentation of a slide and summed across a l l slides in each sequence. (It should be remembered that only the f i r s t four fixations were coded, whereas the total number of fixations was obtained by summing the number of fixations for the f u l l 3 seconds observation time per slide). A two-way analysis of variance (expertise x sequence) with repeated measures on the last factor indicated that there were no significant differences between the groups i n the average number of fixations per slide nor were there any significant two-way interactions involving expertise and sequence. Overall, these findings suggest that the three groups' eye fixations were similar across the six sequences. The low mean was observed for the novices in the unevens, 5.8 fixations per slide; the high was 7.6 per slide for the elites in the beamwalk. The overall mean was 6.7 fixations per slide. There was a significant effect due to sequence with fewer fixations observed in the unevens and more in the walkover and the beamwalk. 50. Phase Analysis The purpose of the phase analysis was to i d e n t i f y differences between the groups across the d i f f e r e n t phases of the move. For t h i s a n a l y s i s , only one sequence was analysed, the beamhand, the longest of the s i x sequences with 26 s l i d e s and the sequence most amenable to segmentation into phases. The sequence was p a r t i t i o n e d into four subsets of s l i d e s s i m i l a r to the four rows of photos shown i n Figure 3. P a r t i t i o n i n g of the movement into hypothesized response u n i t s was done i n consultation with a number of both t h e o r e t i c a l and applied resources and within any constraints imposed by both the number and type of s l i d e s (George, 1980; Carter, 1978). Four temporal units were i d e n t i f i e d : phase 1 -cartwheel to handstand; phase 2 - handstand, quarter turn; phase 3 - the s p l i t s , and phase 4 - the step out. The two beginning s l i d e s were removed to make a t o t a l of 24 and an equal number of s l i d e s i n each phase. Frequency counts were generated for each of the s i x body segments s i m i l a r to the t o t a l sequence a n a l y s i s . Figure 4 shows the mean number of f i x a t i o n s to each segment i n each of the four phases of the beamhand. Insert Figure 4 about here 51. Of i n t e r e s t was whether the groups displayed s i m i l a r or d i f f e r e n t trends i n how they f i x a t e d each body segment across the four temporal phases of the move. Summary of r e s u l t s . The e l i t e subjects d i f f e r e d from the intermediates/novices i n the way they observed the head i n phase two and the hips i n phases two and four of the beamhand (see Table 17). The novices d i f f e r e d from the intermediates i n observation patterns to the shoulders and legs. Across the four phases of the move, the intermediates showed increasing attention to the shoulders with an opposite l i n e a r trend for the novices; f o r the legs (see Figure 4), the intermediates and novices showed d i f f e r e n t cubic trends. The phase analysis also pointed out that the preset orthogonal contrasts had e s s e n t i a l l y "hidden" differences between the groups i n observation to the legs. In Figure 4, i t i s c l e a r that the intermediates not only observed the legs more than the other groups, but t h e i r way of observing the legs was d i s t i n c t i v e l y d i f f e r e n t across the phases i n comparison to the e l i t e s and novices. As i s evident, the phase analysis provided information over and above the t o t a l sequence a n a l y s i s . Results i n d e t a i l . Trend analyses were performed using a set of orthogonal polynomial c o e f f i c i e n t s i n order to determine the extent 52. to which a set of scores follow a l i n e a r (increasing or decreasing), quadratic (U or inverted U), or cubic (N or inverted N) trend. With four phases, the maximum number of trends over which d i r e c t i o n a l change can be assessed i s three. In the phase an a l y s i s , as i n the t o t a l sequence a n a l y s i s , the same set of orthogonal contrasts was used. In the E-IN contrast, the e l i t e subjects were compared to the intermediates and novices combined and i n the I-N contrast, the intermediates were then compared to the novices. There were no contrasts made to test the e f f e c t of f i x a t i o n as a l l four f i x a t i o n s were combined. This was necessary i n order to increase the number of data points thus more adequately t e s t i n g the phase x body segment i n t e r a c t i o n . The mean numbers of f i x a t i o n s to each of the six body segments for each of the four phases are shown i n Table 13. Insert Tables 13 and 14 about here Table 14 presents the three polynomial components observed across the four phases of the move f o r the si x body segment categories by the three groups. The data of Table 14 was further transformed into a 3 (expertise) x 4 (phase trend) contrast e f f e c t s matrix as shown by Table 15. Insert Tables 15 and 16 about here ^ Table 16 presents the multivariate F values f o r the phase x body segment i n t e r a c t i o n i r r e s p e c t i v e of the l e v e l of expertise. M u l t i v a r i a t e l i n e a r , quadratic and cubic trends are shown to be s i g n i f i c a n t . F(6,22) = 5.75, £ .001; F(6,22) = 7.99, £ .001; F(6,22) = 3.53, £ .01, re s p e c t i v e l y . Univariate analyses indicated which of the body segments contributed to these r e s u l t s . S i g n i f i c a n t l i n e a r trends were observed f o r the arms F(l,27) = 7.17 and offbody, F(l,27) = 18.07. Figure 4 shows that, i r r e s p e c t i v e of expertise, the subjects a l l o c a t e d l e s s a ttention to the arms across the phases of the move (negative l i n e a r ) whereas they attended more to offbody elements as the move progressed ( p o s i t i v e l i n e a r ) . Quadratic trends were observed f o r the head, F(l,27) = 26.20 and the hips F(l,27) = 5.41. The three groups attended more to the head during phases one and four and le a s t during the mid phases two and three, i n an inverted U trend. Observations to the hip were i n the reverse d i r e c t i o n with more att e n t i o n a l l o c a t e d to the hip i n phases two and three and les s i n phases one and four. The subjects focussed t h e i r a t t e n t i o n to the head, F(l,27) = 7.59, the arms, F(l,27) = 7.97 and the offbody F(l,27) = 7.07 i n a s i g n i f i c a n t cubic trend, which presents an a d d i t i o n a l feature to the trends already i d e n t i f i e d . I t i s noteworthy that the cubic form of the intermediates i s to the head and arms while the novices scan to the offbody. F i n a l l y , there were no s i g n i f i c a n t trends observed i n the three groups' 54. a l l o c a t i o n of attention to the shoulders or the legs across the phases of the move. Level of expertise by phase by body i n t e r a c t i o n . M u l t i v a r i a t e and univariate F values t e s t i n g the hypotheses E-IN and I-N are shown i n Table 17. The column l a b e l l e d "constant" provides information s i m i l a r to the analyses performed e a r l i e r (see the beamhand i n Table 9). The findings here r e i t e r a t e that the e l i t e s d i f f e r from the intermediates and novices i n t h e i r a l l o c a t i o n of attention to the head, hips and offbody, F(l,27) = 5.77; F(l,27) = 8.07; F(l,27) = 3.95, r e s p e c t i v e l y . The intermediates d i f f e r e d from the novices i n t h e i r a l l o c a t i o n of a t t e n t i o n to the offbody, F(l,27) = 12.73. The l i n e a r , quadratic and cubic trend analyses now take us ins i d e the phases of the move so as to follow up these findings and explain whether there are s i g n i f i c a n t differences between the groups i n the type of trends they ascribe to these body segments across the four phases of the move. No multivariate trends were found to be s i g n i f i c a n t , however, the E-IN u n i v a r i a t e contrast showed that the intermediate/novice subjects d i f f e r e d i n t h e i r Insert Table 17 about here 55. observations to the head, F(l,27) = 6.92, and to the hip with these di f f e r e n c e s occurring i n phase two of the move. In the I-N contrast, the intermediates d i f f e r e d s i g n i f i c a n t l y from the novices i n t h e i r scan patterns to the shoulders across the phases of the move, F(l,27) = 5.03. Both groups displayed l i n e a r trends i n t h e i r observations to the shoulders, however, i n opposite d i r e c t i o n s . The intermediates focused more attention to the shoulders from phase one to four. In contrast, the novices a l l o c a t e d decreasing amounts of att e n t i o n to the shoulders as the move progressed. The two groups displayed d i f f e r e n t cubic p r o f i l e s i n t h e i r a l l o c a t i o n of attention to the legs, F(l,27) = 5.51. Relative to the intermediates, the novices attended most to the legs i n phases two and four and le a s t i n phases one and three. The intermediates were markedly d i f f e r e n t as i s evident i n Figure 4, a l l o c a t i n g s i g n i f i c a n t l y greater amounts of attention to the legs during the f i r s t and t h i r d phases of the move and l e s s during phases two and four than the novices. B. Analysis of the resequencing performance The e l i t e subjects were expected to be both f a s t e r and more accurate i n resequencing the photos of each sequence than the intermediates and novices; the intermediates were expected to be both f a s t e r and more accurate than the novices. As w e l l , the greater the degree of d i f f i c u l t y of the sequence, the greater the l i k e l i h o o d of s i g n i f i c a n t d ifferences between the groups. Two contrasts with respect to the s i x resequencing tasks were made. The f l o o r moves were considered to be of a lower degree of d i f f i c u l t y than the non f l o o r moves; within the apparatus events, the unevens were considered the most d i f f i c u l t of a l l the events, (George, 1980). Summary. The e l i t e gymnasts were both f a s t e r and more accurate than the intermediates, who were i n turn were f a s t e r and committed fewer errors than the novices. It was also confirmed that the more d i f f i c u l t the move, the greater the time needed and the more errors committed. The e l i t e s , intermediates and novices recorded time and error scores that were l i n e a r l y incremental i n a step function that i s r e a d i l y evident i n Figures 5 and 6. The one exception i s the time and error scores i n the v a u l t . Results i n d e t a i l . Figure 5 presents a histogram of the mean time per photo (sec) taken to reconstruct each of the s i x sequences by the three groups. Figure 6 presents the mean errors per photo made during the reconstruction of each sequence. Mean time and error Insert Figures 5 and 6 about here 57. scores on the s i x resequencing tasks are also shown for the three groups i n Table 18. The data of Table 18 were further transformed into a 3 (constant, E-IN, I-N) x 6 ( a l l sequence, and 5 other) contrast matrix, whose estimated e f f e c t s are presented i n Table 19. M u l t i v a r i a t e tests were conducted on time and error measures with the s i x sequences c o n s t i t u t i n g a repeated measures fa c t o r . Univariate analyses were also performed to i d e n t i f y the sources of s i g n i f i c a n t test r e s u l t s . The e f f e c t of l e v e l of expertise was tested by the contrasts E-IN and I-N as defined and used e a r l i e r . The e f f e c t of the degree of d i f f i c u l t y was assessed by f i v e orthogonal contrasts: f l o o r versus apparatus, unevens versus beam and v a u l t , the walkover versus handspring, beamwalk versus beamhand and beam versus v a u l t . The e f f e c t of l e v e l of expertise on a l l sequencing tasks. M u l t i v a r i a t e and uni v a r i a t e F values obtained by the 3 x 6 contrasts matrix analyses are presented i n Table 20. Insert Tables 18 and 19 about here Insert Table 20 about here 58. When the performance of the e l i t e subjects was compared to that of the intermediates and novices, both multivariate and univariate e f f e c t s were found to be s i g n i f i c a n t , M u l t i v a r i a t e F(2,26) = 12.44, p_<.001; Univariate F(1.27) = 10.59 for time and F(l,27) = 7.19 for error. The performance of the intermediates was found to be both f a s t e r and more accurate than that of the novices, M u l t i v a r i a t e F(2,26) = 7.93, £<.001; Univariate F(l,27) = 6.93 for time, F(l,27) = 4.42 for error . The r e s u l t s were c l e a r l y i n l i n e with expectations. The e l i t e subjects were both s i g n i f i c a n t l y f a s t e r and committed fewer errors than the intermediates and novices. In turn, the intermediates were f a s t e r and committed fewer errors than the novices. The e f f e c t s of l e v e l of expertise and degree of d i f f i c u l t y of the  sequences. Table 20 also presents the multivariate and univariate F values corresponding to f i v e contrasts to determine the degree of d i f f i c u l t y of the s i x resequencing tasks. The two f l o o r moves were found to be s i g n i f i c a n t l y d i f f e r e n t , with the walkover leading to more errors than the handspring move, M u l t i v a r i a t e F(2,26) = 3.55, £ .04; Univariate F(l,27) = 4.77 for the error score. The subjects took s i g n i f i c a n t l y l e s s time to resequence the f l o o r moves than to resequence the apparatus moves and t h i s d i f f e r e n c e was not a function of expertise but rather i t was due to the inherent nature 59. of the tasks, M u l t i v a r i a t e F(2,26) = 3.97, £^.03; Univariate F(l,27) = 8.02 for time. As expected, the subjects took s i g n i f i c a n t l y more time and made more errors i n resequencing the unevens than the other apparatus sequences, M u l t i v a r i a t e F(2,26) = 3.36, £^.05; Univariate F(l,27) = 4.20 and 6.62, f o r time and error, r e s p e c t i v e l y . When the expertise l e v e l s were considered, only the novices made s i g n i f i c a n t l y more errors than the intermediates, F(l,27) = 9.45. The beamwalk move took s i g n i f i c a n t l y l e s s time, but led to more errors than the beamhand move. M u l t i v a r i a t e F(2,26) = 5.55, £^.001; Univariate F(l,27) = 4.40 and 5.36 f o r time and error scores, r e s p e c t i v e l y . F i n a l l y , the contrast of the two beam moves with the vault was s i g n i f i c a n t with the beam moves requiring more time to resequence. M u l t i v a r i a t e F(2,26) = 9.34; £<.0001; Univariate F(l,27) = 17.90, £<.0001 f or time. However, the e l i t e subjects took le s s time than the other two groups, Univariate F(l,27) = 4.65; and the intermediates took le s s time than the novices, F(l,27) = 4.75. C. Relationship between eye movements and resequencing a b i l i t y It has been shown e a r l i e r conceptually and/or em p i r i c a l l y that both the subject's eye f i x a t i o n s to the v i s u a l s t i m u l i were a function of t h e i r varying l e v e l s of expertise i n gymnastics. Since s i g n i f i c a n t differences i n these responses were assumed to be based 60. on the d i f f e r e n t types of knowledge of gymnastics, i t i s c r i t i c a l l y important to demonstrate that the two seemingly d i f f e r e n t domains of responses are rel a t e d to one another. To explore the v a l i d i t y of a r e c i p r o c a l function, i t was decided to determine a p r e d i c t i v e v a l i d i t y of the completion time required to reconstruct a sequence of s l i d e s as a function of subjects' eye f i x a t i o n s to c e r t a i n body segments. Since i t was shown that the subjects' f i x a t i o n to the head, hip and offbody d i f f e r e n t i a t e d the e l i t e from the other two groups, i t was decided to explore the r e l a t i o n s h i p between the eye f i x a t i o n s to these body segments ( f i r s t glance and second glance) and the resequencing scores. In order to explore t h i s question, a c o r r e l a t i o n matrix was determined, which involved the resequencing scores and the s i x eye f i x a t i o n v a r i a b l e s , as seen i n Table 21. Insert Table 21 about here ___^§JL_V=__^5:_VVL A multiple regression analysis was performed on the c o r r e l a t i o n matrix, as can be seen i n Table 21, by t r e a t i n g time as the predicted v a r i a b l e and the s i x eye f i x a t i o n v a r i a b l e s as pr e d i c t o r s . The observed multiple c o r r e l a t i o n was .658 which was s i g n i f i c a n t , p_^.029. The empirical regression equation determined i s given i n Table 21 as w e l l . Approximately 43.3% of the variance i n the regression task performance (time) can be explained by a l i n e a r combination of s i x eye v a r i a b l e s , more s i g n i f i c a n t l y by head 2 and hip 2 and offbody 1, the f i r s t and l a s t being most i n f l u e n t i a l , p's^.006 and (.014. The pattern apparent among the s i x regression c o e f f i c i e n t s appears to ind i c a t e a very i n t r i g u i n g p r e d i c t i o n r e l a t i o n . I f the subject's f i x a t i o n patterns were s i m i l a r i n the eye movements and resequencing studies, then one could speculate that the subject's p r o f i c i e n c y i n resequencing the beamhand can be accounted for by the way they a l l o c a t e d a ttention to the head, hips and offbody; the e l i t e attend very l i t t l e to offbody elements or the p o s i t i o n of the head while f i x a t i n g the action of the hip instead. In contrast, the intermediates and e s p e c i a l l y the novices look to offbody elements (background, equipment, movement space) and the o r i e n t a t i o n of the head to guide t h e i r resequencing behaviors. D. Congruence between the introspection reports and eye movements  Summary. Two i n t e r e s t i n g conclusions were derived from t h i s part of the study. F i r s t , the groups were very s i m i l a r i n the incidence of congruences of eye movements to introspections of key points of concentration i n each sequence. Second, the groups d i f f e r e d i n the emphasis they placed on the body segments. The e l i t e subjects were lower body oriented (63.5%); the intermediates were upper body oriented (73.9%); and the novices 60% upper and 40% lower. Results i n d e t a i l . The purpose of the i n t r o s p e c t i o n analysis was to investigate whether there are was any r e l a t i o n s h i p between the athlete's introspection about which body segments contributed most to a successful performance and t h e i r actual eye movements. The experimental procedure f o r t h i s task required the subject to do two things: (1) i d e n t i f y the photograph within the sequence on which the greatest amount of attention would be focussed were she performing the move, and (2) i d e n t i f y the body segments within that s l i d e f e l t to be most important to a successful performance. I t was hypothesized that the groups would d i f f e r i n t h e i r i d e n t i f i c a t i o n of body segments i n t h i s task. In constructing a 3 (expertise) x 6 (sequence) x 5 (body segment) table of congruence instances, as shown i n Appendix V, only the subject's f i r s t s e l e c t i o n of a s l i d e was used as t h i s was considered the most important of t h e i r introspections. The body segments selected (up to four) within t h i s s l i d e were then compared to the subjects' actual focus of a t t e n t i o n as picked up i n the eye movement study. I f at l e a s t one match was found, then a score of 1 was recorded; i f no match was found, then a score of 0. The percent congruence for the e l i t e s was 68.3%, for the intermediates, 76.3%, and f o r the novices 66.6%, average congruence was, 70.4%. A 3 (expertise) x 6 (sequences) x 5 (bodypart) contingency table, as shown i n Table 22, was derived from Appendix V. It shows 63. the congruence between what the subjects observed and what they introspected about with the body segments further categorized into two larger categories, upper body and lower body, (refer to Table 1). Insert Table 22 about here -__SsjL.ks .^i__uja Of p a r t i c u l a r i n t e r e s t here i s the d i f f e r e n t emphasis placed on the body by the three groups. The e l i t e subjects place 36.5% of t h e i r a ttention (whether eye movements or introspection) to the upper body and 63.5% to the lower body. The intermediates are i n the opposite d i r e c t i o n , they a l l o c a t e d 73.9% of t h e i r attention to the upper body and 26.1% to the lower. The novices a l l o c a t e d 60% to the upper body and 40% to the lower, a more in-between stance. A l o g - l i n e a r chi-square analysis was performed on the contingency data i n Table 22. The r e s u l t s of the analysis are presented i n Table 23. Insert Table 23 about here _-^^kr«-ii__\Ni3__ The component l i k e l i h o o d chi-square value of 13.325 c l e a r l y shows that the i n t e r a c t i o n between the contrast of E-IN and the contrast of upper versus lower body part i s s i g n i f i c a n t i n terms of congruence, p_^.01, as described above. Also shown i s the 64. s i g n i f i c a n t i n t e r a c t i o n between the contrast of I-N and that of the upper versus lower body part, c h i square value 13.95, p_(.01. This means that the e l i t e s a l l o c a t e t h e i r attention more congruently to the lower body than do the intermediates/novices. S i m i l a r l y , the intermediates a l l o c a t e t h e i r attention more congruently to the upper body than do the novices to the lower body. 65. IV. Summary of Results, Discussion, and Conclusion A. Summary of r e s u l t s Following a b r i e f summary of the r e s u l t s , the findings w i l l be discussed as a t t r i b u t e s of the gymnast's i n t e r n a l representation or knowledge structure f o r the movements portrayed. In the t o t a l sequence analysis of eye movements, differences between the e l i t e and intermediate/novice gymnasts were confirmed by multivariate analyses i n two of the s i x sequences and by univ a r i a t e analyses of separate body segments i n four of the s i x sequences (see Tables 9 and 10). In comparing the e l i t e gymnasts to the intermediates/novices, the former tended to observe the hips i n the beamwalk and beamhand, and the arms i n the v a u l t . The intermediates/novices attended more to the head i n a l l the sequences except the handspring and to the offbody i n the beamhand. The groups did not d i f f e r i n t h e i r perception of the shoulders and legs. In comparing the intermediate gymnasts to the novices, multivariate analyses revealed differences i n three of the s i x sequences with u n i v a r i a t e analyses of body segments revealing differences between the groups i n f i v e sequences. The intermediates focussed t h e i r a t t e n t i o n more to the head i n a l l the sequences, except the walkover and beamwalk, whereas the novices 6 6 . empha s i z ed o f f b o d y f a c t o r s i n a l l t h e sequences e x c e p t t h e w a l k o v e r and h a n d s p r i n g . The i n t e r m e d i a t e s and n o v i c e s d i d n o t d i f f e r i n t h e i r o b s e r v a t i o n o f t h e h i p . O v e r a l l , d i f f e r e n c e s be tween t h e g r o u p s were l e a s t i n t h e w a l k o v e r and h a n d s p r i n g . Two r e a s o n s a r e o f f e r e d f o r t h i s f i n d i n g : a) t h e y we re t e c h n i c a l l y t h e l e a s t d i f f i c u l t o f t h e s i x s e q u e n c e s , and b ) some o f t h e n o v i c e s c o u l d p e r f o r m t h e s e moves. The g r oup s d i d n o t d i f f e r i n t h e i r p a t t e r n o f f i x a t i o n s f r o m t h e f i r s t t o t h e s econd g l a n c e ( t h e f i r s t two f i x a t i o n s c o m b i n e d , t h e l a s t two f i x a t i o n s c o m b i n e d ) . N e i t h e r d i d t h e y d i f f e r i n t h e a v e r a g e number o f f i x a t i o n s p e r s l i d e . I t was e x p e c t e d t h a t t he e l i t e a t h l e t e s w o u l d r e c o r d f e w e r f i x a t i o n s p e r s l i d e , a r e s u l t r e p o r t e d i n e a r l i e r s t u d i e s ( B a r d and F l e u r y , 1 9 7 6 ) . However , a l l t h r e e g r oup s a v e r a g e d a s i m i l a r 6.7 f i x a t i o n s p e r s l i d e . I n t h e phase a n a l y s i s o f t h e eye movements, t h e d i f f e r e n c e s o b s e r v e d be tween t h e g r oup s i n t h e t o t a l sequence a n a l y s i s we re f u r t h e r e x p l a i n e d r e l a t i v e t o t h e t e m p o r a l pha se s o f t h e move. I n t h e beamhand t o t a l sequence a n a l y s i s , t h e e l i t e s were f ound t o d i f f e r f r o m t h e i n t e r m e d i a t e s / n o v i c e s , i n t h e way t h e y a l l o c a t e d t h e i r a t t e n t i o n t o t h e head and h i p . The phase a n a l y s i s d e t e r m i n e d t h a t t h e s e d i f f e r e n c e s o c c u r r e d i n phase two o f t h e move. The phase a n a l y s i s was a l s o a b l e t o d e t e c t d i f f e r e n c e s be tween t h e g r oup s t h a t were e s s e n t i a l l y " h i d d e n " by t h e t o t a l s equence  a n a l y s i s . F o r e xamp le , i n t h e t o t a l sequence a n a l y s i s t h e 67. i n t e r m e d i a t e s d i d n o t d i f f e r f r o m t h e n o v i c e s i n t h e way th e y v i e w e d t h e s h o u l d e r s and l e g s , however, i n t h e phase a n a l y s i s i t was found t h a t t h e i n t e r m e d i a t e s and n o v i c e s d i s p l a y e d o p p o s i t e t r e n d s i n how th e y o b s e r v e d t h e s e body segments. The i n t e r m e d i a t e s f o c u s s e d more a t t e n t i o n t o t h e s h o u l d e r s as t h e move p r o g r e s s e d , t h e n o v i c e s l e s s ; t h e i n t e r m e d i a t e s a l l o c a t e d g r e a t e r amounts o f a t t e n t i o n t o t h e l e g s d u r i n g phases one and t h r e e and l e s s t o phases two and f o u r , t h e n o v i c e s r e s ponded i n an o p p o s i t e d i r e c t i o n . The r e s u l t s f r o m t h e r e s e q u e n c i n g s t u d y c o n f i r m e d t h e d i s t i n c t i v e n e s s o f t h e group s and l e n t a d d i t i o n a l s u p p o r t t o t h e o v e r a l l r e s e a r c h t h r u s t . The e l i t e gymnasts were b o t h f a s t e r and more a c c u r a t e t h a n t h e i n t e r m e d i a t e s i n r e c o n s t r u c t i n g t h e g y m n a s t i c sequences; t h e i n t e r m e d i a t e s i n t u r n were b o t h f a s t e r and committed f e w e r e r r o r s t h a n t h e n o v i c e s . I t was a l s o c o n f i r m e d t h a t t h e more d i f f i c u l t t h e g y m n a s t i c move, t h e more ti m e r e q u i r e d t o r e s e q u e n c e and t h e g r e a t e r t h e number o f e r r o r s committed. The e l i t e s , i n t e r m e d i a t e s and n o v i c e s r e c o r d e d t i m e and e r r o r s c o r e s t h a t were l i n e a r l y i n c r e m e n t a l i n a s t e p f u n c t i o n , t h a t i s r e a d i l y e v i d e n t i n F i g u r e s 4 and 5. (The one e x c e p t i o n i s t h e t i m e and e r r o r s c o r e s f o r t h e i n t e r m e d i a t e s and n o v i c e s i n t h e v a u l t . ) The f i n a l two a s p e c t s o f t h e r e s e a r c h a) e x p l o r e d t h e r e l a t i o n s h i p between t h e s u b j e c t ' s r e s e q u e n c i n g t i m e and e r r o r s c o r e s and eye f i x a t i o n s ( f o r o n l y t h e beamhand w h i c h was t h e 68. longest of the sequences), and b) the i n t r o s p e c t i o n reports of the gymnasts were compared to t h e i r eye f i x a t i o n s i n a l l the sequences. A multiple regression analysis was performed using the resequencing times a s ^ h e predicted v a r i a b l e and the s i x eye f i x a t i o n scores that distinguished the groups as the p r e d i c t o r s , i . e . , f i r s t glance and second glance f i x a t i o n scores to the head, hip and offbody. The predictor v a r i a b l e s accounted f o r 43.3% of the variance i n the resequencing time scores. In r e l a t i n g the i n t r o s p e c t i o n reports to the eye movements of the subjects, the groups were very s i m i l a r i n incidence of congruences of eye movements to introspections. Approximately 70% of the subject's eye movements and introspections were found to match. However, within that 70%, the groups d i f f e r e d i n the emphasis they placed upon body segments. The e l i t e subjects were lower body oriented (63.5%) i n both eye f i x a t i o n s and introspections, whereas the intermediates were upper body oriented (73.9%) and the novices were 60% upper and 40% lower. These r e s u l t s were consistent with the differences observed i n the eye movements t o t a l sequence analysis of the beamhand where the e l i t e gymnasts were hip and centre body oriented, the intermediates/ novices more head and arms oriented. In summary, the r e s u l t s were c l e a r i n showing that the groups d i f f e r e d c o n s i s t e n t l y when asked to observe gymnastic s k i l l s or 69. solve problems with or introspect about stimulus material from gymnastics. B. Discussion The c l e a r l y d i f f e r e n t focus of att e n t i o n of the groups i n the eye movement study, the l i n e a r l y incremental time and error scores i n the resequencing task, the r e s u l t s i n the regression analysis and the int r o s p e c t i o n reports, i n combination, lend considerable support to the idea that gymnasts possess i n t e r n a l representations or knowledge structures f o r the movements i n t h e i r sport that d i f f e r as a function of s k i l l l e v e l . More importantly, the three studies suggest that a measure of understanding of these knowledge structures can be gained by measuring how they observe the performance of others, or solve problems using stimulus information from gymnastics. Bernstein (1967) and Turvey (1977) portray the a c q u i s i t i o n of s k i l l as the gaining of con t r o l over the degrees of freedom of a movement problem. That i s , as the performer evolves from novice to e l i t e status, the type of cognitive c o n t r o l they exercise over t h e i r muscle system changes such that co-ordinated actions r e s u l t from " r e l a t i v e l y simple" executive i n s t r u c t i o n s . To date, i t has been d i f f i c u l t to determine how the executive l e v e l stategy of an athlete changes as s k i l l develops. The r e s u l t s i n t h i s study suggest that by systematically studying the perceptual input of the 70. a t h l e t e , r e l a t i v e to stimulus information from t h e i r sport, may evolve "perceptual i n v a r i a n t s " that are stable and a function of a p a r t i c u l a r l e v e l of expertise. Evidence was presented showing that observers are perceptually s e n s i t i v e to pictures portraying d i f f e r e n t orientations or movements of the human body and that t h i s s e n s i t i v i t y i s rel a t e d to t h e i r a b i l i t y to both mentally and p h y s i c a l l y model or perform the portrayed movements (Sekiyama, 1982, 1983; Freyd, 1983). Second, when an observer views the actions of others, they discriminate or make decisions by e x t r i c a t i n g perceptual i n v a r i a n t s (Johannson, 1979; Cutting and Kozlowski, 1978; Todd, 1983). Perceptual inv a r i a n t s are elements i n a changing event that p e r s i s t or are perceived as being invariant and are fundamental to the process of i d e n t i f i c a t i o n , d i s c r i m i n a t i o n and judgement. Cutting (1978) described the "center of moment" as being an invar i a n t of gender i d e n t i f i c a t i o n and Todd (1983) defined the action of the lower leg as the perceptual invariant that distinguished running from walking. Warren (1983) showed that how we perceive a p a r t i c u l a r human action may be rela t e d to our own ph y s i c a l s i z e . In the eye movements study, the three groups perceived elements of the action i n a way that was unique or in v a r i a n t f o r t h e i r l e v e l of expertise i n gymnastics. Are these, then, perceptual i n v a r i a n t s r e l a t e d to s k i l l l e v el? For example, i n the beam moves the e l i t e gymnasts were hip or centre of the body oriented, a p o t e n t i a l perceptual invariant not displayed by the other groups. This f i n d i n g i s i n contrast to the intermediates, who were drawn to observe the head, and the novices, the offbody f a c t o r s . Several questions a r i s e . Are novice athletes so drawn to the surrounding factors i n the performance environment that u n t i l they are attuned to these aspects they are simply unable to perceive other elements? Why i n the beam moves are the intermediates attuned to observe the head and arms of the athlete and not the hip actio n , a d i r e c t contrast to the e l i t e performers? Perhaps the e l i t e gymnasts are aware of the centre of the body as being the key to e s t a b l i s h i n g consistent performance i n t h i s event. Perceptual attunement to the "center of moment" may be a perceptual invariant of the beam where balance i s so c r u c i a l and gymnasts become co g n i t i v e l y attuned to t h i s as a function of s k i l l attainment. The other groups did not give perceptual p r i o r i t y to the centre region of the body as they did not r e a l i z e i t s importance to performance i n the event. The introspection study added support to the d i f f e r e n t i a l focus of attention by the gymnasts of d i f f e r e n t s k i l l l e v e l s . Across a l l the sequences, the e l i t e gymnasts reportedly placed a greater emphasis upon the lower body while the intermediates placed greater emphasis on the upper body. The novice were reported to a l l o c a t e more attention to the upper body, a tendency s i m i l a r to the intermediates' focus of a t t e n t i o n . Is t h i s because the intermediates, i n p a r t i c u l a r , are s t i l l s truggling with f l e x i b i l i t y and power generation as w e l l as safety factors that are perceived to reside i n the upper body? These are the types of questions that would be pursued i n defining perceptual invariants of gymnastic acti o n . With the detection and d e l i n e a t i o n of information such as t h i s w i l l come the eventual d e s c r i p t i o n of knowledge structures of actions. A d i s t i n c t i o n was made between conceptually driven and data driven perceptual systems. The r e s u l t s of the resequencing study established that the a b i l i t y to solve problems using v i s u a l information from gymnastics i s a relevant a t t r i b u t e of the knowledge structure of action and appeared to be linked to s k i l l l e v e l . The r e s u l t s indicated that the e l i t e gymnasts brought something "extra" to bear i n reconstructing the sequences as exhibited by t h e i r e f f i c i e n t , e r r o r - f r e e reconstruction of the photographs. When presented with the out-of-order frames of movements i n gymnastics, they appear to have been guided by a deeper understanding of each s k i l l than what was provided by the surface features of the photographs. This would seem to imply an i n t e r n a l knowledge structure f o r the movement that was r e l a t e d to t h e i r experience i n performing the movement. Because t h e i r s k i l l l e v e l was more re f i n e d and at a higher l e v e l than both the other groups, t h e i r perceptual-problem solving processes with gymnastic 73. information was also more re f i n e d and e f f i c i e n t . The same appeared to be the case f o r the intermediates when compared to the novices. The "fragment theory" of r e t r i e v a l process proposed by Jones, (1976) and Broadbent, (1981) provides an analo g i c a l basis f o r explaining differences between the groups on the resequencing task. Fragment theory holds that we b u i l d i n t e r n a l representations i n long-term memory composed of fragments, each of which may contain several components of the event. If one such component i s l a t e r used as a probe, i t allows r e c a l l of the other components i n that fragment. By t h i s d e f i n i t i o n then, the e l i t e athletes would be viewed as having incorporated a vast array of "fragments" about gymnastics into a cohesive plan or knowledge structure for gymnastics that was more f i n e l y structured than e i t h e r of the other l e s s e r s k i l l e d groups. One might speculate that when confronted with the photographic "fragments" of a gymnastic sequence, they perceived not feature dependent s t i m u l i but the t o t a l move. In summary, the observed differences between the three groups may be explained i n terms of the a p p l i c a t i o n of perceptual/ cognitive structures to various forms of input from gymnastics. The groups of gymnasts were viewed as possessing i n t e r n a l representations or knowledge structures f o r the s k i l l s of gymnastics that were accessed through the use of eye movement, resequencing and in t r o s p e c t i o n tasks. Concepts such as perceptual i n v a r i a n t s and conceptually driven versus data driven processes 74. provided u s e f u l tools for describing cognitive differences between the gymnasts of d i f f e r e n t s k i l l l e v e l s . Further pursuit of the expert-novice dimension i n terms of empirical v a r i a b l e s would lead to a more complete d e s c r i p t i o n of the gymnast's knowledge structure of action. C. Uniqueness and implications of the research findings The r e s u l t s i n the three studies lend support to a research framework that seeks to understand differences between athletes i n terms of the knowledge they possess about t h e i r sport. The present approach placed an emphasis upon the information pick-up of the athlete f o r the purpose of understanding what i t i s i n informational terms that distinguishes one s k i l l l e v e l from another. The approach, therefore, r e l i e d not upon measures of ph y s i c a l behavior but on measures of the information the athlete may be using when executing these behaviors. An approach such as t h i s represents what Kuhn (1970) has c a l l e d a picking up of the other end of the s t i c k , that i s , a look at an old problem but from another angle. The present research strategy brought into the laboratory s e t t i n g athletes with d i f f e r i n g l e v e l s of mastery of sport s k i l l s developed out of the laboratory. Thus, i t provided the advantage of analyzing the knowledge structures of athletes involved i n the " r e a l " world. As w e l l , the information they were asked to deal with was taken from the sport i t s e l f , through the use of f i l m s , photographs and s l i d e s . This approach i s both a strength and a weakness of the research strategy. The strength l i e s i n the type of information derived, i . e . , the r e s u l t s speak d i r e c t l y to events that occur i n the everyday world of the a t h l e t e . For example, the information derived i n the three studies here was for s p e c i f i c s k i l l s i n gymnastics. This information i s now a v a i l a b l e f o r use by i n d i v i d u a l s , e s p e c i a l l y teachers and coaches, i n that sport. As we l l , the research methods used here are applicable to most a c t i v i t i e s and to a wide range of events within those a c t i v i t i e s . Subject groups of a l l types may be used—young, o l d , s k i l l e d , n o n - s k i l l e d , male, female, s p e c i a l groups i n a range of roles -athlete, coach, teacher, judge, sportscaster, analyst. The weakness of the approach i s that, although the findings do speak to " r e a l " events i n " r e a l " sports they are confined to the p a r t i c u l a r stimulus material used. In t h i s study, only s i x sequences were used and the r e s u l t s obtained are r e l a t i v e to those sequences. This drawback i s p a r t i a l l y overcome when one considers that a l l research i s i n i t i a l l y task dependent and only achieves a range of g e n e r a l i z a b i l i t y a f t e r extensive and varied use of a p a r t i c u l a r task. The u t i l i z a t i o n of " r e a l i s t i c " s t i m u l i i s also more e a s i l y defended when one considers research p r a c t i s e s i n many other f i e l d s today. For example, research i n language, reading and speech comprehension i s no longer conducted using semantically barren 76. l e t t e r s or even words; instead s c r i p t s are used with a l l t h e i r richness, complexity and meaning. Expert-novice research uses stimulus information derived from the domain being s t u d i e d — problems i n physics, x-rays of lungs with and without malignancy, computer languages basic to the science; s i m i l a r l y t y p i s t s are studied as they type paragraphs and p i a n i s t s play Bach. When stimulus information i s used that i s too s i m p l i s t i c and not a f a i r expression of the knowledge that characterizes a f i e l d , the r e s u l t s are equally l i m i t e d and constrained. So too i n sport; the knowledge an athlete possesses i s exceptionally complex and i t i s only when t h i s information i s introduced into our research that we w i l l come to understand how or i n what manner i t resides within the athlete. D. Some suggested applications of the research f i n d i n g s .  Contribution to future research Measures such as perceptual i n v a r i a n t s , resequencing time and error norms, introspection reports plus many others could be derived for d i f f e r e n t s k i l l l e v e l s ( e l i t e , intermediate, or novice) and i n d i f f e r e n t roles (athletes, coaches, teachers, judges, analysts, and so forth) i n any number of a c t i v i t i e s . The development of cognitive p r o f i l e s that accompany s k i l l attainment might p a r a l l e l aerobic or anaerobic norms for p h y s i o l o g i c a l function. Instead of measures of oxygen consumption, we would possess i n d i c a t o r s of information consumption. Contribution to teaching and coaching r o l e s . The development of t h i s type of information would bring about an awareness of the athlete, teacher and coach that we presently do not have. For example, i f i t were established that the centering of the performer's attention to the hip region of the body i s a c h a r a c t e r i s t i c of " e l i t i s m " i n gymnastics, t h i s f i n d i n g would have an impact upon a) the screening or early detection of p o t e n t i a l e l i t e l e v e l gymnasts, b) the development of teaching and t r a i n i n g methods designed to f a c i l i t a t e such awareness, and c) the detection of errors and the diagnosis and evaluation of performance. Also, a perennial debate i n p r o f e s s i o n a l preparation programs i s whether the p h y s i c a l s k i l l l e v e l of the student teacher i s a factor i n subsequent teaching and coaching r o l e s . The e f f e c t of p r i o r s k i l l attainment upon subsequent teaching and r o l e s cannot be answered by t h i s study. However, the r e s u l t s do suggest that the knowledge that athletes possess about t h e i r sport d i f f e r s with s k i l l l e v e l . A next stage of the research would involve expert and novice coaches d i f f e r e n t i a t e d by s k i l l l e v e l , thus t e s t i n g for the i n t e r a c t i o n of p r i o r experience on coaching. As w e l l , these coaching r e s u l t s would be compared to the a t h l e t e findings i n t h i s study, thereby assessing the extent to which an athlete's knowledge 78. structure remains s t a t i c or changes as a function of assuming teaching and coaching r o l e s . Another i n t e r e s t i n g question i s to explore the extent to which a performer's report of events i n a contest a c t u a l l y coincide with an objective assessment of that event. Franks (1983; personal conversation, 1983), has developed objective methods f or quantifying the events i n a soccer match, i . e . , turnovers, possessions, corners, shots on goal, etc. He has found very l i t t l e r e l a t i o n s h i p between these objective r e s u l t s and what coaches v e r b a l l y describe or r e c a l l l a t e r . I f a coach's perceptions of the game d i f f e r from r e a l i t y , i t becomes important to determine what information the coach does consider most important. The methods used i n t h i s study are a step toward i d e n t i f y i n g the information pick-up of the coach and others involved i n sport. This type of information can be extremely important since what the coach perceives to have occurred tends to serve as the basis f o r t h e i r d ecision making and actions i n preparing game plans, s u b s t i t u t i n g players or a l t e r i n g t a c t i c s . S i m i l a r l y , the same could be said f o r the athlete and a l l others who are involved i n sporting contexts. Contribution to i n s t r u c t i o n a l design. The information that constitutes a d e s c r i p t i o n of knowledge structures of actions could have an impact on the design of i n s t r u c t i o n . The evolution of knowledge structures as portrayed i n t h i s study i s d i s t i n c t from what i s a v a i l a b l e now as they are derived d i r e c t l y from the learner rather than being i n t e r p r e t a t i v e , that i s , developed by teachers and coaches who are removed i n a sense from the event. Our i n s t r u c t i o n a l resources are currently l a r g e l y i n t e r p r e t i v e (Vickers, 1983); the existence of information that i s obtained d i r e c t l y from the athlete permits a more e x p l i c i t recognition of the types of information that are held to be most meaningful at t h e i r l e v e l . E. Conclusion In summary, the aims of the present i n v e s t i g a t i o n were achieved. Three cognitive studies detected perceptual, problem solving and i n t r o s p e c t i v e differences among groups of gymnasts of d i f f e r i n g a b i l i t i e s . The differences of multi-faceted nature, were taken as contributing to the i d e n t i f i c a t i o n of relevant a t t r i b u t e s of the knowledge structure of the gymnasts. The implications of the findings were further examined i n terms of the notion of the c e n t r a l representations of a c t i o n , the r o l e of v i s u a l perception i n tapping t h i s representation, the i d e n t i f i c a t i o n of perceptual invariants as w e l l as the d i s t i n c t i o n between conceptually driven and data driven perceptual systems. The r e s u l t s achieved were described as a t t r i b u t e s of the gymnast's knowledge structure with a p p l i c a t i o n s to research, teaching, coaching and the design of i n s t r u c t i o n . 80. References Anderson, J.R. (1980). Cognitive psychology and i t s implications. San Francisco: W.H. Freeman and Co. Antes, J.R. (1974). The time course of p i c t u r e viewing. Journal of Experimental Psychology, 103, 62-70. Antes, J.R. & Penland, J.G. (1981). Picture context e f f e c t s on eye movement patterns. In D.F. Fisher, R.A. Monty, & J.W. Senders (Eds.), Eye movements: cognition and v i s u a l perception (pp. 157-170). H i l l s d a l e , N.J.: Lawrence Erlbaum. Bard, C. & Fleury, M. (1976). Analysis of v i s u a l search a c t i v i t y during sport problem s i t u a t i o n s . Journal of Human Movement Studies, 3, 214-222. Bard, C. & Fleury, M. (1981). Considering eye movement as a p r e d i c t i o n of attainment. In I.M. C o c k e r i l l & W.W. MacGillivray (Eds.), V i s i o n and sport: Cheltenham: Stanley Thornes Publishers (Ltd). Bard, C , Fleury, M., C a r r i e r e , L. & Halle, M. (1980). Analysis of gymnastic judges' v i s u a l search, Research Quarterly, 51, (2), 267-273. B a r t l e t t , F.C. (1932). Remembering: A study i n experimental and  s o c i a l psychology. Cambridge: Cambridge Uni v e r s i t y Press. 81. Bernstein, N. (1967). The co-ordination and regulation of movements. Oxford: Pergamon Press. Broadbent, D.E. (1981). Association l e c t u r e : From the percept to the cognitive structure. In J. Long & A. Baddeley (Eds.), Attention and Performance IX. H i l l s d a l e , New Jersey: Lawrence Erlbaum Associates.. Carter, E. (1978). Gymnastics f o r g i r l s and women. Englewood C l i f f s , N.J.: Prentice H a l l . Chase, W.G. & Simon, H. (1973). Perception i n Chess. Cognitive Psychology, 4, 55-81. Chi, M.T.H. & Glaser, R. (1980). The measurement of expertise: Analysis of the development of knowledge and s k i l l as a basis for assessing achievement. In E.L. Baker & E.S. Quellmalz (Eds.), Educational t e s t i n g and evaluation. Beverly H i l l s , CA: Sage Publications. Chi, M.T.H., Glaser, R. & Rees, E. (1982). Expertise i n problem solving. In R. Steinberg (Ed.) Advances i n the psychology of human i n t e l l i g e n c e . H i l l s d a l e , N.J.: Lawrence Earlbaum. Cummings, G.D. (1978). Eye movements and v i s u a l perception. In E.C. Catherette & M.P. Friedman (Eds.), Handbook of perception, (Vol. 9, pp. 221-255). New York: Academic Press. Cutting, J.E. (1978). Generation of synthetic male and female walkers through manipulation of a biomechanical i n v a r i a n t . Perception, ]_, 393-405. 82. Cutting, J.E. & Kozlowski, L.T. (1977). Recognizing friends by t h e i r walk: Gait perception without f a m i l i a r i t y cues. B u l l e t i n of the Psychonomic Society, £(5), 353-356. Cutting, J.E. & P r o f f i t t , D.R. (1981). Gait perception as an example of how we may perceive events. In R.D. Walk & H.L. Pick (Eds.), Intersensory perception and sensory i n t e g r a t i o n, (pp. 249-273). New York: Plenum Pub1. Corp. Cutting, J.E., P r o f f i t t , D.R. & Kozlowski, L. (1978). A biomechanical invariant f o r g a i t perception. Journal of Experimental Psychology: Human Perception and Performance, 4(3), 357-372. Donald, J.G. (1980). Structures of knowledge and implications f o r  teaching. (Report #6). Vancouver: University of B r i t i s h Columbia, Faculty of Education, Centre f o r the Improvement of Teaching. Franks, I., Goodman, D., & M i l l e r , G. (1983). Analysis of performance: q u a l i t a t i v e or qu a n t i t a t i v e . (Science P e r i o d i c a l  on Research and Technology i n Sport). Ottawa: Coaching Association of Canada. Freyd, J . J . (1983). The mental representation of movement when s t a t i c s t i m u l i are viewed. Perception and Psychophysics, 33(6), 575-581. 83. Friedman, A. & L i e b e l t , L.S. (1981). On the time course of viewing pictures with a view towards remembering. In D.F. Fisher, R.A. Monty & J.W. Senders, (Eds.) Eye movements: Cognition and v i s u a l perception. H i l l s d a l e , N.J.: Lawrence Erlbaum Associates. G a l l i s t e l , CR. (1980). The organization of ac t i o n . A new synthesis. H i l l s d a l e , N.J.: Lawrence Erlbaum Associates. George, G.S. (1980). Biomechanics of women's gymnastics, Englewood C l i f f s , N.J.: Prentice H a l l . Gibson, J . J . (1979). The e c o l o g i c a l approach to v i s u a l perception. Boston: Houghton M i f f l i n . Glaser, R. (1976). Components of a psychology of i n s t r u c t i o n : Toward a science of design, Review of educational psychology, 46(1), 1-24. Glencross, D.J. (1978). Psychology and sport. Sydney: McGraw H i l l . Haber, R.N. (1983). The impending demise of the icon: A c r i t i q u e of the concept of i c o n i c storage i n v i s u a l information processing. The behavioural and b r a i n sciences, (>(1), 1-54. Hainline, L. (1981). Eye movements and form perception i n human in f a n t s . In D.F. Fisher, R.A. Monty & J.W. Senders, Eye  movements: Cognition and v i s u a l perception (pp. 3-19). H i l l s d a l e , N.J.: Lawrence Erlbaum Associates. 84. Hunt, M. (1982). The universe within. A new science explores the  human mind. Brighton, Sussex: The Harvester Press. Johnson, D.G., Severance & P.J. F e l t o v i c h . (1979). Design of deci s i o n support systems i n medicine. In Proceedings, 12th  Hawaii International Conference on System Sciences, (pp. 105-118). Johannson, G. (1975). V i s u a l motion perception. S c i e n t i f i c American, 232, 67-75. Johannson, G. (1977). S p a t i a l constancy and motion i n v i s u a l perception. In W. Epstein (Ed.), S t a b i l i t y and constancy i n v i s u a l perception: Mechanisms and processes. New York: John Wiley & Sons. Johannson, G. (1979). Memory functions i n v i s u a l event perception. In L.G. Nils s o n , (Ed.) Perspectives on memory  research, essays i n honor of Uppsala University's 500th  anniversary (pp. 93-101). New York: John Wiley & Sons. Johannson, G. (1983). Optic flow, icons, and memory. Commentary/Haber: Goodbye to the icon. The Behavioural and  Brain Sciences, 6(1), 23-24. Jones, G.V. (1976). A fragmentation hypothesis of memory: Cued r e c a l l of pictures and of sequential p o s i t i o n . Journal of Experimental Psychology: General, 105, 277-293. 85. Kozlowski, L.T. & Cutting, J.E. (1977). Recognizing the sex of a walker from a dynamic p o i n t - l i g h t display. Perception and  Psychophysics, 2_1(6) , 575-580. Kuhn, T.S. (1970). The structure of s c i e n t i f i c revolutions. Chicago: Un i v e r s i t y of Chicago Press. Kundel, H.L. & Nodine, C F . (1978). Studies of eye movements and v i s u a l search i n radiology. In J.W. Senders, D.F. Fisher & R.A. Monty, (Eds.) Eye movements and the higher psychological  functions, (pp. 317-328). H i l l s d a l e , N.J.: Lawrence Erlbaum Associates. Lee, D.N. (1978). The function of v i s i o n . In H. Pick and E. Salzmann (Eds.), Modes of perceiving and processing  information, (pp. 159-170). H i l l s d a l e , N.J.: Lawrence Erlbaum Associates. Lee, D.N. & Aronson, E. (1974). V i s u a l proprioceptive control of standing i n human i n f a n t s . Perception and Psychophysics, 15, 527-532. Lindsay, P.H. & Norman, D.A. (1977). Human information processing (2nd ed.). New York: Academic Press. Loftus, G.R. (1982). Picture memory methodology. In Richard Huff (Ed.), Handbook of research methods i n human memory and cognition. New York: Academic Press. 86. Loftus, G.R. & Mackworth, N.H. (1978). Cognitive determinants of f i x a t i o n l o c a t i o n during p i c t u r e viewing. Journal of Experimental Psychology: Human Perception and Performance, 4_, 565-572. Mackworth, N.H. & Morandi, A.J. (1967). The gaze s e l e c t s informative d e t a i l s within p i c t u r e s . Perception and Psychophysics, 2, 547-551. Mace, M.M. (1983). Proceedings of a meeting of the i n t e r n a t i o n a l society of e c o l o g i c a l psychology. Journal of Experimental  Psychology: Human Perception and Performance, j ) ( l ) , 151-157. McKeithen, K., Reitman, J . , Rueter, H. & H i r t l e , S. (1981). Knowledge organization and s k i l l d ifferences i n computer programmers. Cognitive Psychology, 13, 307-325. Minsky, M. (1975). A framework f o r representing knowledge. In P.H. Winston (Ed.), The psychology of computer v i s i o n , (pp. 211-277). New York: McGraw H i l l . Neisser, U. (1976). Cognition and R e a l i t y . San Francisco: Freeman. Neisser, U. (1983). The r i s e and f a l l of the sensory r e g i s t e r . Commentary/Haber: Goodbye to the icon. The behavioural and bra i n sciences, 6, 35. 87. Newell, K.M. & Barclay, CR. (1982). Developing knowledge about acti o n . In J.A.S. Kelso & J.E. Clarke (Eds.), The development  of movement con t r o l and co-ordination (pp. 175-212). Toronto: John Wiley & Sons. Norman, D.A. (1981). A psychologist views human processing: Human errors and other phenomenon suggest processing mechanisms. In Proceedings of the 7th I.J.C.A.I., LI, 1097-1101. Norman, D.A. (1982). Learning and memory. San Francisco: W.H. Freeman and Co. Noton, D. & Stark, L. (1971). Eye movements and v i s u a l perception. S c i e n t i f i c American, 224, 34-43. Pew, R.W. (1966). A c q u i s i t i o n of h i e r a r c h i c a l control over the temporal organization of a s k i l l . Journal of Experimental  Psychology, 71, 764-771. Pew, R.W. (1974). Human perceptual performance. In B.J. Kantowitz (Ed.), Human Information Processing: T u t o r i a l s i n  performance and cognition. H i l l s d a l e , N.J.: Lawrence Erlbaum Associates. Reed, E.S. (1982). An o u t l i n e of a theory of a c t i o n systems. Journal of Motor Behavior, L4(2), 98-134. Rumelhart, D., Lindsay, P. & Norman, D. (1972). A process model fo r long-term memory. In E. Tulving and W. Donaldson (Eds.), Organization of memory, New York: Academic Press. Rumelhart, D.G. & Ortony, A. (1977). The representation of knowledge i n memory. In R.C. Anderson, R.J. Spiro and W.E. Montague (Eds.), Schooling and the a c q u i s i t i o n of knowledge. H i l l s d a l e , N.J.: Lawrence Erlbaum Associates. Rumelhart, D.E. & Norman, D.A. (1981). Analogical processes i n learning. In J.A. Anderson (Ed.), Cognitive s k i l l s and t h e i r a c q u i s i t i o n . H i l l s d a l e , N.J.: Lawrence Erlbaum Associates. Runeson, S. (1977). On the p o s s i b i l i t y of "smart" perceptual mechanisms. Scandinavian Journal of Psychology, 18, 172-179. Runeson, S. & Frykholm, G. (1981). V i s u a l perception of l i f t e d weight. Journal of Experimental Psychology: Human Perception and Performance, _7, 733-740. Schmidt, R.A. (1975). A schema theory of di s c r e t e motor s k i l l l earning. Psychological Review, 82, 225-260. Schmidt, R.A. (1982). Motor c o n t r o l and learning: A behavioral emphasis. I l l i n o i s : Human K i n e t i c s . Sekiyama, K. (1982). Kinesthetic e f f e c t s of mental representations i n the i d e n t i f i c a t i o n of l e f t and r i g h t hands. Perception and Psychophysics, 32(2), 89-95. Sekiyama, K. (1983). Mental and p h y s i c a l movements of hands: Kine s t h e t i c information preserved i n representational systems. Japanese Psychological Research, 25 (2), 95-102. Shaffer, L.H. (1981). Performances of Chopin, Bach and Bartok: Studies i n motor programming. Cognitive Psychology, 13, 326-278. 89. Shank, R.C. and Abelson, R. (1977). S c r i p t s , plans, goals and  understanding: An inquiry into human knowledge structures. H i l l s d a l e , N.J.: Lawrence Erlbaum Associates. Sheena, D. & Flagg, N. (1978). Semiautomatic eye movement data analysis techniques f o r experiments with varying scenes. In J.W. Senders, D.F. Fisher & R.A. Monty (Eds.), Eye movements  and the higher psychological functions. H i l l s d a l e , N.J.: Lawrence Erlbaum Associates. Simon, H.A. (1980). Invited address to the American Educational Research Association. Simon, H.A. (1981). The sciences of the a r t i f i c i a l (2nd ed.). Cambridge: The MIT Press. Simon, H.A. (1979). Models of thought. New Haven: Yale Uni v e r s i t y Press. Stark, L. & E l l i s , S.R. (1981). Scanpaths r e v i s i t e d : Cognitive models d i r e c t a c t i v e looking. In D.F. Fisher, J.N. Senders & R.A. Monty, (Eds.) Eye movements: Cognition and v i s u a l  perception (pp. 193-226). H i l l s d a l e , N.J.: Lawrence Erlbaum Associates. Todd, J.T. (1983). Perception of g a i t . Journal of Experimental  Psychology: Human Perception and Performance, 9_(1), 31-42. Turvey, M.T. (1973). A note on the r e l a t i o n between action and perception, i n M.G. Wade and R. Martens (Eds.), Psychology of  motor behavior and sport, (pp. 307-313). Proceedings of the NASPSPA. Urbana, I l l i n o i s : Human K i n e t i c s Publ. 90. Turvey, M.T. (1977a). Preliminaries to a theory of action with reference to v i s i o n . In R. Shaw & J . Bransford (Eds.), Perceiving, acting, and knowing; Toward an e c o l o g i c a l  psychology, (pp. 211-265). H i l l s d a l e , N.J.: Lawrence Erlbaum Associates. Turvey, M.T. (1977b). Contrasting orientations to a theory of v i s u a l information processing. Psychological Review, 84, 67-88. Vickers, J.N. (1983). The ro l e of expert knowledge structures i n an I n s t r u c t i o n a l Design Model for Physical Education. Journal  of Teaching i n Physical Education, 2_(3) , 17-32. Whiting, H.T.A. (1980). Dimensions of cont r o l i n motor learning. In. G.E. Stelmach & J . Requin (Eds.), T u t o r i a l s i n motor  behavior (pp. 537-550). North Holland Publ. Co. Wickelgren, W. (1981). Human learning and memory. Annual Review  of Psychology, 32, 21-52. Women's Gymnastics, (1982). [ f i l m s e r i e s ] . North Palm Beach, F l o r i d a : The A t h l e t i c I n s t i t u t e . Young, L.R. & Sheena, D. (1975). Survey of eye movement recording methods. Behavior Research Methods and Instrumentation, _7(5), 397-429. Yarbus, A.L. (1967). Eye movements and v i s i o n . New York: Plenum Press. Zernicke, R.F., Gregor, R.J. & Cratty, B.J. (1982). Balance and v i s u a l proprioception i n ch i l d r e n . Journal of Human Movement  Studies, 8, pp. 1-13. Table 1 Three Levels of Coding as Applied to Eye Movement Data i n Six Movement Sequences i n Gymnastics  Name of Sequence Walk- Hand Beam Beam over Spring Unevens Walk Hand Vault Number of Sl i d e s (22) (13) (18) (22) (26) (17) 2-Category 6-Category 18 Category Coding Coding Coding 01 - Upper 01 - Head 01 - Head X X X X X X Body 02 - Shoulder- •02 - Shoulder X X X X X X 03 - Arms. 03 - Upper Arms X X X X X X - 04 - Fore Arm X X X X X X >05 - Hand X X X X X X -06 - Upper Torso X X X X X X "07 - Lower Torso X X X X X X 02 - Lower 04 - Hips 0^8 - Hips X X X X X X Body 05 - Legs •09 - Upper Leg Leading X X X X 10 - Lower Leg Leading X X X X 11 - Upper Leg T r a i l i n g X X X X 12 - Lower Leg T r a i l i n g X X X X 13 - Upper Legs X X X 14 - Knees X X X Ll5 - Lower Legs/Feet X X X 06 - Offbody rl6 - Equipment X X X X X X • 17 - Background X X X X X X Ll8 - Movement Space X X X X X X Number of categories used (x) 15 15 14 18 18 14 Table 2 Eye Movements T o t a l Sequence Analysis f o r Forward Walkover Observed Mean 3 x 2 Contrast E f f e c t s f o r Six Body Segments A l l Four Fixations F i r s t vs. Second Glance Body Segment Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody Constant 8.00**15.70** 8.13**15.07**29.13**10.50** 0.87* 3.37**-0.27 1.27* -3.07**-1.43 E-IN a -4.20**-3.30 0.25 2.30 4.30 0.45 -1.00 -1.90 0.55 0.20 1.90 0.95 I-N 2.60 -1.40 -0.10 -0.40 3.80 -3.30 2.20* -0.60 2.90* -3.00* -0.20 -0.70 MSerror 14.84 38.89 11.30 26.44 79.78 31.84 3.36 13.08 6.51 9.13 24.50 9.22 E - e l i t e gymnasts; I -** p <.00 * p <.05 intermediate gymnasts; N - novice gymnasts. Table 3 Eye Movements T o t a l Sequence Analysis for Forward Handspring Observed Mean 3 x 2 Contrast E f f e c t s f o r Six Body Segments A l l Four Fixations F i r s t vs. Second Glance Body Segment Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody Constant 5.60** 9.30** 4.17** 8.93**15.33** 6.40** 0.20 2.30**-0.97** 1.27**-l.87**-0.40 E-IN 3 -1.50 1.05 1.25 1.15 -2.60 0.30 -0.6 -0.15 -1.25 1.25 -0.50 0.60 I-N 2.40* -1.50 0.50 -0.90 2.00 -2.60 1.80* -0.30 1.90* -0.70 -1.40 -0.80 MSerror 6.79 13.92 5.65 14.70 25.10 22.62 2.53 10.88 3.28 5.15 11.56 7.39 E- e l i t e gymnasts; I - intermediate gymnasts; N - novice gymnasts ** p (.00 * p <.05 Table 4 Eye Movements Total Sequence Analysis for the Uneven Bars Observed Mean 3 x 2 Contrast Effects for Six Body Segments A l l Four Fixations First vs. Second Glance Bosy Segment Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody Constant 8.20** 5.57**14.70** 6.17**16.90**14.23** -0.33 0.83* 1.03 -0.63 -0.03 0.03 E-INa -2.40 0.95 -0.3 1.70 0.60 -2.45 0.80 -0.65 1.60 -1.0 -1.0 -0.65 I-N 3.20* 3.70* 1.80 2.80 0.80 -13.90** -1.20 0.10 0.80 0.20 2.00 -1.10 MSerror 10.04 10.33 20.43 10.66 24.26 28.93 4.12 3.83 16.54 5.19 5.27 5.78 E - el i t e gymnasts; I - intermediate gymnasts; N - novice gymnasts ** p <.00 * p (.05 Table 5 Eye Movements Total Sequence Analysis f o r Beam Back Walkover Observed Mean 3 x 2 Contrast E f f e c t s f o r Six Body Segments A l l Four Fixations F i r s t vs. Second Glance Body Segment Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody Constant 6.83**14.60** 9.30**15.50**28.73**11.00** -0.03 1.13 -0.70 3.17** 0.87 -2.40** E-IN 3 -4.10** 0.90 -0.30, 6.60**-0.95 -1.65 0.50 0.70 0.30 2.30* -3.25 -1.05 I-N 1.0 -5.20 3.20 -1.60 10.70* -7.90* 0.20 -2.40 -0.40 1.80 -0.30 0.90 MSerror 13.97 46.84 16.98 28.60 91.98 46.29 6.56 14.13 8.48 8.69 24.47 7.92 E - e l i t e gymnasts; I - intermediate gymnasts; N - novice gymnasts ** p <.00 * p <.05 Table 6 Eye Movements To t a l Sequence Analysis f o r Beam Handstand Observed Mean 3 x 2 Contrast E f f e c t s f o r Six Body Segments A l l Four Fixations F i r s t vs. Second Glance Body Segment Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody Constant 8.87**17.07**13.80**15.93**32.73**14.07** -0.47 1.00 -0.67 2.33** 0.73 -2.20** E-IN a -3.70* 3.20 1.65 7.75**-1.85 -6.55* 1.30 0.30 0.55 2.35* -3.95 -1.05 I-N 2.40 -0.80 3.50 -0.50 8.30 -13.30** 0.40 -1.0 -1.90 2.30 -2.30 1.30 MSerror 13.24 36.31 28.42 38.00 101.50 60.13 8.64 12.24 14.91 7.46 37.53 9.67 E - e l i t e gymnasts; I - intermediate gymnasts; N - novice gymnasts ** p (.00 * p <.05 Table 7 Eye Movements T o t a l Sequence Analysis f o r Vault Observed Mean 3 x 2 Contrast E f f e c t s for Six Body Segments A l l Four Fixations F i r s t vs. Second Glance Body Segment Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody Constant 10.47**11.07** 9.50**10.33**11.70** 8.70** 1.60**-0.93 -0.43 -1.80** 0.23 1.30* E-IN a -2.95* -1.00 2.10* 1.75 0.90 -2.25 0.15 1.70 0.50 -2.25* -0.80 0.75 I-N 2.70 1.40 1.00 1.90 2.40 -7.30** 0.50 -0.40 0.20 -0.10 -0.60 0.30 MSerror 9.67 11.31 6.11 9.41 14.89 18.45 6.36 7.62 10.94 8.04 11.83 10.30 E - e l i t e gymnasts; I - intermediate gymnasts; N - novice gymnasts ** p <.00 * p <.05 Table 8 Eye Movements Tot a l Sequence Analysis: Description of Constant Contrast E f f e c t s by Six Body Segments P r o f i l e s of Body Segments Ir r e s p e c t i v e of Expertise and Fi x a t i o n Levels Sequence Walkover Handspring Unevens Beam-Walk Beam-Hand Vault Contrast F Value P F Value P F Value P F Value P F Value P F Value P M u l t i v a r i a t e 6144.14 .00 1776.07 .00 1074.84 .00 13202.62 .00 4558.73 .00 2386.70 .00 Univariate HE 129.41 .00 138.50 .00 200.83 .00 100.30 .00 178.18 .00 339.99 .00 SH 190.17 .00 186.47 .00 90.00 .00 136.53 .00 240.65 .00 324.83 .00 AR 175.68 .00 92.21 .00 317.38 .00 152.80 .00 201.01 .00 442.78 .00 HP 257.60 .00 162.83 .00 107.06 .00 252.00 .00 200.39 .00 340.24 .00 LG 319.17 .00 281.05 .00 353.14 .00 269.28 .00 316.68 .00 275.75 .00 OB 103.88 .00 54.32 .00 210.03 .00 78.42 .00 98.73 .00 123.09 .00 Table 9 Eye Movements T o t a l Sequence Analysis: Testing of Two Expertise Level Contrasts Effects by Six Body Segments Expertise x Body Segment Int e r a c t i o n Irrespective of Fixation Levels Sequence Walkover Handspring Unevens Beam -Walk Beam -Hand Vault Contrast F Value p F Value p F Value p F Value p F Value p F Value p E-IN a M u l t i v a r i a t e 1 .56 .21 .99 .45 1 .77 .15 2.20 .08 2 .85 .03** 2 .83 .03* Univariate HE 7 .95 .00** 2 .21 .15 3 .82 .06 8.02 .00** 6 .89 .01* 6 .00 .02* SH 1 .87 .18 .53 .47 .58 .45 .11 .74 1 .88 .18 .59 .45 AR .04 .85 1 .84 .18 .03 .86 .03 .85 .64 .43 4 .80 .04* HP 1 .33 .26 .60 .45 1 .81 .19 10.15 .00** 10 .54 .00** 2 .17 .15 LG 1 .55 .22 1 .80 .19 .10 .76 .07 .80 .22 .64 .36 .55 OB .04 .84 .03 .87 1 .38 .25 .39 .54 4 .76 .04* 1 .83 .19 I-N 3 M u l t i v a r i a t e .88 .53 1 .32 .29 6 .28 .00** 1.89 .13 2 .74 .04* 3 .98 .00** Univariate HE 2 .28 .14 4 .24 .05* 5 .09 .03* .36 .55 2 .18 .15 3 .77 .06 SH .25 .62 .81 .38 6 .63 .02* 2.89 .10 .09 .77 .87 .36 AR .00 .95 .22 .64 .79 .38 3.02 .09 2 .16 .15 .82 .38 HP .03 .86 .28 .60 3 .68 .07 .45 .51 .03 .85 1 .92 .17 LG .90 .35 .80 .38 .13 .72 6.22 .02* 3 .39 .07 1 .93 .17 OB 1 .71 .20 1 .49 .23 33 .38 .00** 6.74 .02* 14 .71 .00** 14 .44 .00** E - e l i t e gymnasts; I - intermediate gymnasts; N - novice gymnasts ** p (.00 * p (.05 Table 10 Eye Movements To t a l Sequence Analysis: Summary of Differences i n Fixations to Body Segments i n Six Sequences i n Gymnastics Irrespective of Fixation Level Body Segment Head Shoulders Arms Sequence 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 ' 4 5 6 E l i t e vs. Intermediates and Novices .00 < .06 < .00 < .01 < .02 < .04 > Intermediates vs. Novices .05 > .03 > .15 > .06 > .02 > .10 < .09 > .15 > Body Segment Hips Legs Offbody Sequence 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6 E l i t e vs. Intermediates and Novices .00 > .00 > .15 > .04 < Intermediates vs. Novices .07 > .02 . > ,07 > .00 < .02 < .00 < .00 < ^ - fewer f i x a t i o n s to; /- more f i x a t i o n s to. 1 = Walkover 2 = Handspring 3 = Unevens 4 = Beam Walkover 5 = Beam Handstand 6 = Vault Table 11 Eye Movements T o t a l Sequence Analysis: Testing of F i r s t vs. Second Glance Contrast E f f e c t s by 6 Body Segments F i r s t vs. Second Glance x Body Segment Interaction Irrespective of Expertise Levels Sequence Walkover Handspring Unevens Beam-Walk Beam--Hand Vault Contrast F Value • P F Value p F Value p F Value P F Value p F Value p Grand Mean Irrespective of Expertise M u l t i v a r i a t e 6.76 .00** 7.03 .00** 2.37 .06 19.80 .00** 14.39 .00** 4.90 .00** Univariate HE 6.70 .01** .48 .50 .81 .38 .00 .94 .76 .39 12.07 .00** SH 26.0 .00** 14.59 .00** 5.46 .03* 2.72 .11 2.45 .13 3.43 .08 AR .32 .57 8.55 .00** 1.93 .18 1.73 .20 .89 .35 .52 .48 HP 5.27 .03** 9.35 .00** 2.32 .14 34.61 .00** 21.89 .00** 12.09 .00** LG 11.51 .00** 9.05 .00** .00 .94 .92 .34 .43 .52 .14 .71 OB 6.68 .02* .65 .43 .00 .94 21.82 .00** 15.02 .00** 4.92 .04* E - e l i t e gymnasts; I - intermediate gymnasts; N - novice gymnasts ** p <.00 * p <.05 a Table 12 Eye Movements Tot a l Sequence Analysis of Variance: Testing of 2 Expertise Level Contrasts E f f e c t s x F i r s t vs. Second Glance by Six Body Segments Expertise x F i x a t i o n : F i r s t vs. Second Glance x Body Segment Interaction Sequence Walkover Handspring Unevens Beam-Walk Beam--Hand Vault Contrast F Value P F Value P F Value P F Value P F Value p F Value P E-IN Mu l t i v a r i a t e .80 .58 .81 .57 1.22 .33 2.16 .08 1.81 .14 .92 .50 Univariate HE 1.98 .17 .95 .34 1.04 .32 .25 .62 1.30 .26 .02 .88 SH 1.84 .19 .01 .90 .74 .40 .23 .63 .04 .83 2.53 .12 AR .31 .58 3.17 .08 1.03 .32 .07 .79 .14 .72 .15 .70 HP .30 .86 2.02 .17 1.28 .27 4.06 .05* 4.90 .03* 4.19 .05* LG .98 .33 .14 .71 1.26 .27 2.88 .10 2.72 .10 .36 .55 OB .65 .43 .32 .49 .49 .93 .34 .76 .39 .36 .55 I-N M u l t i v a r i a t e 2.19 .08 1.93 .12 .78 .60 .55 .76 1.77 .15 .05 .99 Univariate HE 7.21 .01* 6.43 .02* 7.20 .19 .03 .86 .09 .76 .20 .66 SH .14 .71 .04 .84 .01 .90 2.04 .16 .40 .53 .11 .75 AR 6.46 .02* 5.51 .02* .19 .66 .09 .76 1.21 .28 .12 .89 HP 4.93 .04* .48 .50 .04 .85 1.86 .18 3.55 .67 .00 .94 LG .00 .93 .85 .37 3.79 .07 .02 .89 .70 .41 .15 .96 OB .26 .61 .43 .52 1.05 .32 .51 .48 .87 .36 .04 .83 E - e l i t e gymnasts; I - intermediate gymnasts; N - novice gymnasts ** p <.00 * p <.05 Table 13 Eye Movement Phase Analysis Mean Number of Fixations to Six Body Segments for Beamhand Phase 1 Phase 2 Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody E l i t e 2.2 4.6 3.7 4.3 6.8 2.3 0.2 5.6 4.0 6.1 6.9 1.3 Intermediate 2.9 2.9 3.3 3.1 9.7 1.9 2.4 3.1 4.8 3.4 8.6 1.3 Novice 2.9 3.8 3.3 3.4 6.0 4.4 1.3 4.8 4.2 3.0 7.0 3.3 Phase 3 Phase 3 Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody E l i t e 0.5 4.9 3.1 4.6 8.1 2.9 2.4 3.3 2.9 4.1 8.3 2.7 Intermediate 0.8 3.3 2.7 4.0 10.7 2.3 3.0 4.9 3.5 2.4 6.6 3.5 Novice 0.6 3.6 2.5 4.1 6.7 6.2 2.6 3.0 1.0 2.3 7.9 6.6 Table 14 Eye Movement Phase A n a l y s i s : Observed Means of Polynomial Component by Six Body Segments Trend Across 4 Phases by 3 Expertise Levels Head Shoulders Arms Constant Linear Quad Cubic Constant Linear Quad Cubic Constant Linear Quad Cubic E l i t e 2.65 0. .20 1.95 -0. .16 9, .20 -1.02 -1.30 0. 18 6, .85 -0.74 -0, .25 0, .42 Intermediate 4.55 -0. .29 1.35 1. .10 7, .10 1.39 0.70 0. 31 7, .15 -0.33 -0. .35 1, .45 Novice 3.70 -0. .36 1.80 0. ,40 7. .60 -0.80 -0.80 0. 63 5. .50 -1.92 -1. ,20 0, .63 Hips Legs Offbody Constant Linear Quad Cubic Constant Linear Quad Cubic Constant Linear Quad Cubic E l i t e 9, .55 -0. 47 -1. .15 0, .96 15, .05 1.27 0.05 -0. .47 4. ,60 0. .63 0. 40 -0. .98 Intermediate 6, .45 -0. 34 -0. .95 -0. .56 17. .80 -1.61 -1.50 -2. .10 4. ,50 1. .30 0. 90 -0, .31 Novice 6, .40 -0. 49 -0. ,70 -0, .98 13. .80 1.2 0.10 0, .63 10, .25 2, .12 0. 75 -1. .45 Table 15 Eye Movements Phase Analysis: Observed 3 Expertise Level Contrasts and 4 Phase Polynomial Trend E f f e c t s by Six Body Segments Head Shoulders Arms Constant Linear Quad Cubic Constant Linear Quad Cubic Constant Linear Quad Cubic Grand Mean 3.63** -0.15 1.70** 0.45* 7.97** 0.15 -0.47 0.37 6.50** -1.00* -0.60 0.83* E-IN a -1.48* 0.53 0.38 -0.91* 1.85i -1.32i -1.25* -0.29 0.53 0.39 0.53 -0.61 I-N 0.85 0.07 -0.45 0.69i -0.50 2.19* 1.50 -0.31 1.65 1.591 0.85 0.83 Hips Legs Offbody Constant Linear Quad Cubic Constant Linear Quad Cubic Constant Linear Quad Cubic Mean 7.47** -0.43 0.93* -0.19 15.55** 0.29 -0.45 -0.65 6.45** 1.35** 0.68i -0.92* E-IN 3.12* -0.06 -0.33 1.73* 0.75 1.48 0.75 0.27 -2.781 -1.08 -0.43 -0.10 I-N 0.05 0.15 -0.25 0.42 4.001 -2.82i -1.60 -2.73* -5.75** -0.83 0.15 1.14 ** p (.001 * p (.05 i p ( .15 o Table 16 Eye Movements Phase Analysis of the Beamhand: Testing of 4 Phase Polynomial Trend Components E f f e c t s by Six Body Segments f o r Grand Mean of Expertise Levels. Phase and Body Segment Interaction Irrespective of Expertise Levels Contrast Constant Linear Quadratic Cubic F Value £ F Value £ F Value £ F Value £ Grand Mean Mu l t i v a r i a t e 8840.00 .00* 5.75 .00** 7.99 .00** 3.53 .01* Univariate Head 157.60 .00* 0.13 .73 26.20 .00** 7.59 .01* Shoulders 193.70 .00* 0.14 .71 1.07 .31 0.93 .34 Arms 190.60 .00* 7.17 .01* 3.06 .09 7.97 .00* Hips 207.29 .00* 1.48 .23 5.41 .03* 0.40 .53 Legs 304.88 .00* 0.23 .63 0.60 .44 1.87 .18 Offbody 96.14 .00* 18.07 .00** 3.35 .08 7.07 .01* Table 17 Eye Movements Phase Analysis of the Beam Hand: Testing of 2 Expertise Level Contrast x Phase Polynomial Trend Components E f f e c t s by Six Body Segments Expertise x Phase x Body Segments Interaction Contrast Constant Linear Quadratic Cubic F Value P F Value P F Value P F Value P E-IN 3 M u l t i v a r i a t e 2.88 .03* 1.18 .35 0.78 .59 2.14 .09 Univariate Head 5.77 .02* .34 .56 0.28 .60 6.92 .01* Shoulders 2.32 .14 2.43 .13 1.70 .20 0.13 .73 Arms .28 .60 0.24 .63 0.52 .48 0.96 .34 Hips 8.07 .00** 0.00 .94 0.15 .71 7.18 .01* Legs .16 .69 1.34 .26 0.37 .55 0.07 .79 Offbody 3.95 .05* 2.59 .12 0.29 .60 0.02 .89 I-N M u l t i v a r i a t e F 2.19 .08 1.30 .30 0.91 .50 1.27 .31 Univariate F Head 1.44 .24 0.00 .95 0.31 .58 3.04 .09 Shoulders .13 .72 5.03 .03* 1.84 .19 0.11 .74 Arms 2.05 .16 3.02 .09 1.02 .32 1.30 .26 Hips .00 .97 0.03 .86 0.06 .80 0.32 .57 Legs 3.36 .08 3.67 .07 1.27 .27 5.50 .03* Offbody 12.73 .00** 1.13 .30 0.03 .87 1.82 .19 E - e l i t e gymnasts; I - intermediate gymnasts; N - novice gymnasts ** p (.00 * p (.05 a Table 18 Analysis of the Resequencing Task: Mean Time and Errors per S l i d e f o r Six Sequences Sequence Walkover Handspring Unevens Beam Walk Beam Hand Vault Expertise Time Erro r s Time Errors Time Errors Time Errors Time Errors Time Errors E l i t e 3.77 .09 3.13 .05 4.08 .05 3.48 .10 3.98 .03 3.19 .04 Intermediate 4.47 .13 4.03 .03 5.63 .11 5.06 .15 6.62 .08 4.79 .12 Novice 7.28 .13 7.05 .09 9.34 .34 8.64 .21 9.25 .14 4.75 .10 Table 19 Analysis of the Resequencing Task: Observed 3 Expertise Level Contrasts and 6 Sequence Contrasts E f f e c t s Walkover vs. Floor vs. Unevens vs. Beam Walk vs. Beam vs. A l l Sequences Handspring Apparatus Beam and Vault Beam Hand Vault Contrast Time Err o r s Time Errors Time Errors Time Errors Time Errors Time Errors Constant 32.84** 7.53** 0.44 0.56* -3.11** -1.63 2.46* 2.43* -0.89* 0.70* 3.86** 0.77 E-IN a --16.82** -5.30* 0.30 -0.25 3.29 3.65 -1.32 -3.65 0.59 0.15 -4.17* -0.25 I-N -15.71* -4.80* 0.22 0.70 -1.79 2.70 -4.99 -7.10** -0.94 0.10 -6.30** -1.19 E - e l i t e gymnasts; I - intermediate gymnasts; N - novice gymnasts ** p ( .00 * p (.05 Table 20 Analysis of the Resequencing Task: Testing of 3 Expertise Level Contrasts x Six Sequence Contrasts E f f e c t s Walkover vs. Floor vs. Unevens vs. Beam Walk vs. Beam vs. A l l Sequences Handspring Apparatus Beam and Vault Beam Hand Vault F Value P F Value P F Value P F Value P F Value P F Value P Constant M u l t i v a r i a t e 168.60 .00** 3.55 .04* 3.97 .03* 3.36 .05* 5.55 .00** 9.34 .00** Univariate Time 181.66 .00** 2.34 .13 8.02 .00** 4.20 .05* 4.40 .05* 17.90 .00** Erro r 65.39 .00** 4.77 .04* 2.78 .11 6.66 .02* 5.36 .03* 2.69 .11 E-IN M u l t i v a r i a t e 12. .44 .00** .21 .81 1.74 .20 1, .90 .17 .22 .81 2 .24 .13 Univariate Time 10. .59 .00** .24 .63 2.00 .17 .27 .61 .43 .52 4 .65 .04* Erro r 7. .19 .01* .21 .65 3.08 .09 3, .33 .08 .05 .82 .06 .80 -N Mu l t i v a r i a t e 7. .93 .00** .64 .53 1.41 .26 4, .58 .02* .43 .66 4 .75 .02* Univariate Time 6, .93 .01* .09 .76 .44 .51 2, .89 .10 .82 .37 7 .95 .00** Error 4. .42 .04* 1.21 .28 1.27 .27 9. .45 .00** .02 .89 2 .75 .11 ** p (.00 * p <.05 Table 21 Means and I n t e r c o r r e l a t i o n Matrix of Resequencing Task Completion Time and Four Selected Eye Fixations Scores (Beamhand) (N=30) Y XI X2 X3 X4 X5 X6 (Time) (Head 1) (Head 2) (Hip 1) (Hip 2) (Obi) (0b2) Mean 79.37 4.20 4.67 9.13 6.80 5.93 8.13 S.D. 30.89 2.21 2.46 3.78 2.91 3.80 4.51 XI -.1439 1.000 X2 .4552 .0623 1.000 X3 -.2151 -.1956 -.3469 1.000 X4 -.1112 -.2539 -.2335 .7260 1.000 X5 .4060 .0305 .0688 -.4885 -.3779 1.000 X6 .2926 -.2925 .2312 -.5017 -.1929 .6788 1.000 Multi p l e R (y,X £) = .658 £ .029 R 2 = .43 Pr e d i c t i o n Equation: y = 46.41 - 4.875 (Head 1) + 6.808 (Head 2) - .622 (Hip 1) (p<.006) + 2.218 (Hip 2 ) + 5.865 (Obyl) - 2.613 (0by2) (P(.014) Legend: Y (Time): Time/sec to complete resequencing task. XI (Head 1): Fixations to Head at f i r s t glance. X2 (Head 2): Fixations to Head at second glance. X3 (Hip 1): Fixations to Hip at f i r s t glance. X4 (Hip 2): Fixations to Hip at second glance. X5 (Obi): Fixations to Offbody at f i r s t glance. X6 (0b2): Fixations to Offbody at second glance. Table 22 Frequency Contingency Table of Introspection Responses Congruent with Eye Fixati o n s to Six Body Segments by Expertise Levels and Six Sequences SEQUENCE WALKOVER HANDSPRING UNEVENS BEAMWALK BEAMHAND VAULT TOTAL % Exper- Con-t i s e gruence Upper Lower Upper Lower Upper Lower Upper Lower Upper Lower Upper Lower Upper Lower Matching 0 6 4 4 2 3 3 6 3 3 3 4 15 26 (36.5) (63.5) E l i t e ' Non- 10 4 6 6 8 7 7 4 7 7 7 6 45 34 Matching Matching 4 2 5 2 4 6 4 2 8 0 8 0 34 12 (73.9) (26.1) Inter-mediate Non- 6 8 5 8 6 4 6 8 2 10 2 10 27 48 Matching Matching 2 6 5 4 2 3 4 1 5 1 6 1 24 16 (60) (40) Novice Non- 8 4 5 6 8 7 6 9 5 9 4 9 36 44 Matching Table 23 Log-Linear Analysis o f a 3 x 6 x 2 x 2 Chi-squares Contingency Table (Table 27) by Orthogonal Sets of Contrasts Cumulative Component Sample Response Likelihood Likelihood Contrasts Contrasts Chi-Squares df. Chi-Squares df. £ Constant Mat-Non 54.546 65 .061 1 E-IN Mat-Non 54.485 64 .383 2 Upp-Low 54.102 62 13.325 2 .01 Mat x Upp 40.773 60 .658 10 Sequences 40.075 50 I-N Mat-Non 52.662 63 .799 3 Upp-Low 51.863 60 13.945 3 .01 Mat-Low 37.918 57 1.068 15 Sequences 36.850 42 X 2 (.01) - 6 ' 6 3 5 ' X 2 ( . 0 1 ) 2 - 9 ' 2 1 0 ' X 2 ( . 0 1 ) 3 = U ' 3 4 5 Figure Caption Figure 1. A conceptual framework for exploring knowledge structures of action. * A t t r i b u t e s of a knowledge structure expressed as units of a n a l y s i s . COGNITIVE TASKS Figure Caption Figure 2. Two examples of eye movement data to a body segment. In photograph 2a, the shoulder i s f i x a t e d , code 02. In photograph 2b, an example of movement space i s shown, code 18. Photograph sequence extracted from Women's Gymnastics, Beginning Level, Class 111 Beam and Beginning Level, Class 111 Floor Exercise, (1982), 16 mm. f i l m produced by the A t h l e t i c I n s t i t u t e , North Palm Beach, Florida.(permission granted). 1 1 7 . 118. Figure Caption Figure 3. The beam handstand sequence showing 12 of the 26 s l i d e s used i n the eye movement study. Photograph sequence extracted from Women's Gymnastics, Beginning Level, Class 111 Beam, (1982), produced by The A t h l e t i c I n s t i t u t e , North Palm Beach, F l o r i d a , (permission granted). 120 . Figure Caption Figure 4. Mean f i x a t i o n s to s i x body segments at four phases of the beam handstand sequence. rH CM MEAN NUMBER OF F I X A T I O N S A 10 — i — P i P2 P3 HEAD P4 MEAN NUMBER OF F I X A T I O N S A 10 H 8 H 2 H LT 3-PI P2 P3 P4 SHOULDER MEAN NUMBER OF F I X A T I O N S A 10 4 1 2 H PI P2 P3 ARM • E L I T E * INTER • NOVICE — i — PA MEAN NUMBER MEAN NUMBER MEAN NUMBER OF F I X A T I O N S OF F I X A T I O N S OF F I X A T I O N S HIP LEGS OFFBODY 122 . Figure Caption Figure 5. Mean time per photograph (sec) i n resequencing s i x movement sequences i n gymnastics. TIME/PHOTO [sec] 10 4 • WALKOVER HANDSPRING UNEVENS • ELITE H INTERMEDIATE • NOVICE GYMNASTIC SEQUENCES BEAMWALK BEAMHAND VAULT F i g u r e C a p t i o n F i g u r e 6. Mean e r r o r s p e r p h o t o g r a p h i n r e s e q u e n c i n g s i x movement sequences i n g y m n a s t i c s . NUMBER OF ERRORS/PHOTO A 0.50 -0.40 0.30 -0.20 -0.10 -Jl WALKOVER HANDSPRING UNEVENS • ELITE S INTERMEDIATE • NOVICE GYMNASTIC SEQUENCES BEAMWALK BEAMHAND VAULT 126. 127. • SI ? 9 m r o m m <Q CD 13 O Q_ > 128. FIXATION ONE FIXATION TWO FIXATION THREE FIXATION FOUR m m 3 a . 130. DBI Z z H 1 o m m _ jo rn C b <Q CD m 3 g Q. > 131. Appendix G FIXATION FOUR FIXATION THREE A M • FIXATION TWO FIXATION ONE »m WI M l fh HEAD SHOULDERS ARMS HIPS LEGS OFFBODY M l HEAD SHOULDERS ARMS HIPS LEGS OFFBODY HEAD SHOULDERS ARMS HIPS LEGS OFFBODY HEAD SHOULDERS ARMS -IQ-BL HIPS LEGS OFFBODY MEAN FIXATIONS TO 6 BODY SEGMENTS - WALKOVER • ELITE H INTERMEDIATE • NOVICE Appendix H FIXATION FOUR FIXATION THREE FIXATION TWO FIXATION ONE *>h T I mn •>[ HEAD SHOULDERS ARMS fi] HIPS LEGS OFFBODY l ^ i l ^ n HEAD SHOULDERS ARMS HIPS LEGS fi] OFFBODY HEAD SHOULDERS ARMS M c i _ J L _ JsD HIPS LEGS OFFBODY mfh II HEAD SHOULDERS ARMS HIPS LEGS OFFBODY MEAN FIXATIONS TO 6 BODY SEGMENTS - HANDSPRING • ELITE H INTERMEDIATE • NOVICE Appendix I FIXATION FOUR FIXATION THREE FIXATION TNO FIXATION ONE b L f i i HEAD SHOULDERS ARMS HIPS LEGS OFFBODY Hi ITI fh Fn HEAD SHOULDERS ARMS HIPS LE6S OFFBODY Hi WFh Bi i l l Hi F HEAD SHOULDERS ARMS HIPS LE6S OFFBODY Hi i n fcn in m BJI HEAD SHOULDERS ARMS HIPS LEGS DFFBDQ O ODY MEAN FIXATIONS TO 6 BODY SEGMENTS - UNEVENS • ELITE H INTERMEDIATE • NOVICE Appendix J FIXATION FOUR FIXATION THREE FIXATION TWO 4 FIXATION ONE i l l g h i f ] in tin HEAD SHOULDERS ARMS HEAD HEAD SHOULDERS ARMS HIPS LEGS OFFBODY I HEAD SHOULDERS ARMS HIPS LEGS OFFBODY HIPS LEGS OFFBODY E gri ^  • ELITE • INTERMEDIATE • NOVICE SHOULDERS ARMS HIPS LEGS OFFBODY MEAN FIXATIONS TO 5 BODY SEGMENTS - BEAM-BACK WALKOVER Appendix K FIXATION FOUR FIXATION THREE FIXATION TWO FIXATION ONE •hi F i th i n rfn BI th Bi in th HEAD SHOULDERS ARMS HIPS LEGS OFFBODY I HEAD SHOULDERS ARMS HIPS LEGS OFFBODY HEAD SHOULDERS ARMS HIPS LEGS OFFBODY HEAD SHOULDERS ARMS HIPS LEGS OFFBODY MEAN FIXATIONS TO 6 BODY SEGMENTS - BEAMHAND • ELITE S INTERMEDIATE • NOVICE Appendix L FIXATION FOUR FIXATION THREE FIXATION TWO FIXATION ONE jktL taJEb HEAD SHOULDERS ARMS HIPS LEGS OFFBODY fil_tn_Jl HEAD SHOULDERS ARMS to HIPS LEGS OFFBODY tffl B n in « - i HEAD SHOULDERS ARMS HIPS LEGS OFFBODY tfl FT1 fcn rf3n fn HEAD SHOULDERS ARMS UTDC HIPS LEGS OFFBODY MEAN FIXATIONS TO 6 BODY SEGMENTS - VAULT • ELITE • INTERMEDIATE • NOVICE Appendix M Eye Movements Total Sequence Analysis f o r Forward Walkover C e l l Mean Fixations to Six Body Segments by Expertise and Fixation Levels F i r s t Glance Second Glance Body Segment Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody E l i t e 2.7 7.8 4.2 9.0 15.1 5.0 2.5 5.7 4.1 7.6 16.9 5.8 Intermediate 6.5 9.9 4.5 6.9 12.9 3.3 " 4.2 6.2 3.5 7.2 16.7 5.4 Novice 4.1 10.9 3.1 8.6 11.1 5.3 4.0 6.6 5.0 5.9 14.7 6.7 S.D. 2.21 3.55 2.17 2.94 4.76 3.13 2.06 3.66 2.05 3.03 5.43 3.27 Appendix N Eye Movements T o t a l Sequence Analysis f o r Forward Handspring C e l l Mean Fixations to Six Body Segments by Expertise and Fixation Levels F i r s t Glance Second Glance Body Segment Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody E l i t e 2.2 6.1 1.6 5.9 5.7 3.3 2.4 3.9. 3.4 3.8 7.9 3.3 Intermediate 4.3 5.2 2.2 4.3 7.4 2.0 3.0 3.0 1.8 3.8 9.8 3.0 Novice 2.2 6.1 1.0 5.1 7.1 3.7 2.7 3.6 2.5 3.9 8.1 3.9 S.D. 1.48 2.69 1.56 2.18 2.62 2.75 1.57 2.27 1.42 2.28 3.39 2.73 Appendix 0 Eye Movements Tot a l Sequence Analysis f o r Uneven Bars C e l l Mean Fixations to Six Body Segments by Expertise and Fixation Levels F i r s t Glance Second Glance Body Segment Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody E l i t e 3.4 3.3 8.3 3.0 8.3 6.1 3.2 2.9 6.2 4.3 9.0 6.5 Intermediate 4.7 4.1 8.3 3.4 9.2 3.9 5.9 3.0 7.4 3.6 7.? 4.2 Novice 3.7 2.2 7.0 1.9 7.8 11.4 3.7 1.2 6.9 2.3 8.5 10.6 S.D. 2.11 1.93 2.76 1.86 2.78 2.79 1.62 1.83 3.29 2.11 2.34 3.09 Appendix P Eye Movements T o t a l Sequence Analysis f o r Beam Back Walkover C e l l Mean Fixation s to Six Body Segments by Expertise and Fixation Levels F i r s t Glance Second Glance Body Segment Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody E l i t e 2.2 8.4 4.3 12.3 13.4 3.4 1.9 6.8 4.8 7.6 14.7 6.5 Intermediate 4.3 5.7 5.0 7.9 18.1 3.0 4.4 6.0 6.0 4.6 16.3 4.6 Novice 3.7 9.5 3.6 7.8 12.9 6.5 4.0 7.4 4.2 6.3 10.8 9.0 S.D. 2.24 4.13 2.63 3.57 4.99 3.18 2.29 3.66 2.41 2.42 5.77 4.12 Appendix Q Eye Movements T o t a l Sequence Analysis f o r Beam Handstand C e l l Mean Fixation s to Six Body Segments by Expertise and Fi x a t i o n Levels F i r s t Glance Second Glance Body Segment Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody E l i t e 3.4 10.2 7.3 12.5 14.8 3.4 3.0 9.0 7.6 8.6 16.7 6.3 Intermediate 5.3 8.0 6.6 7.9 19.2 4.2 6.0 7.6 8.4 5.2 18.3 5.4 Novice 3.9 8.9 5.8 7.0 16.2 10.2 5.0 7.5 5.7 6.6 13.0 12.7 S.D. 2.21 3.57 3.53 '3.78 5.93 3.80 2.46 3.39 3.02 2.91 5.86 4.51 Appendix R Eye Movements T o t a l Sequence Analysis for Vault C e l l Mean Fixations to Six Body Segments by Expertise and Fixation Levels F i r s t Glance Second Glance Body Segment Head Shoulder Arm Hip Leg Offbody Head Shoulder Arm Hip Leg Offbody E l i t e 5.1 5.3 5.4 4.1 6.0 4.5 3.4 5.1 5.5 7.4 6.3 2.7 Intermediate 7.3 5.2 4.4 4.8 6.4 3.5 5.5 6.9 4.9 5.9 6.2 2.3 Novice 5.7 4.7 3.8 3.9 5.5 5.0 4.4 6.0 4.5 4.9 4.7 6.1 S.D. 1.99 2.54 2.23 1.98 2.89 3.06 2.01 1.74 1.88 2.20 2.21 2.24 Appendix S Eye Movements Tot a l Sequence Analysis - A l l Four Fixations Combined Sequence Forward Walkover Forward Handspring Body Segment Head Shoulder Arms Hips Legs Offbody Head Shoulder Arms Hips Legs Offbody E l i t e 5.2 13.5 8.3 16.6 32.0 10.8 4.6 10.0 5.0" -9.7 13.6 6.6 Intermediate 10.7 16.1 8.0 14.1 30.0 8.7 7.3 8.2 4.0 8.1 17.2 5.0 Novice 8.1 17.5 8.1 14.5 25.8 12.0. 4.9 9.7 3.5 9.0 15.2 7.6 S.D. 3.9 6.2 3.4 5.1 8.9 5.6 2.6 3.7 2.4 3.8 5.9 4.8 Appendix T Eye Movements T o t a l Sequence Analysis - A l l Four Fixations Combined Sequence Uneven Bars Beam Back Walkover Body Segment Head Shoulder Arms Hips Legs Offbody Head Shoulder Arms Hips Legs Offbody E l i t e 6.6 6.2 14.5 7.3 17.3 12.6 4.1 15.2 9.1 19.9 28.1 9.9 Intermediate 10.6 7.1 15.7 7.0 17.1 8.1 8.7 11.7 11.0 12.5 34.4 7.6 Novice 7.4 3.4 13.9 4.2 16.3 22.0 7.7 16.9 7.8 14.1 23.7 15.5 S.D. 3.2 3.2 4.5 3.0 4.5 5.4 3.7 6.8 4.1 5.3 9.6 6.8 Appendix U Eye Movements Tot a l Sequence Analysis - A l l Four Fixations Combined Sequence Forward Walkover Forward Handspring Body Segment Head Shoulder Arms Hips Legs Offbody Head Shoulder Arms Hips Legs Offbody E l i t e 6.4 19.2 14.9 21.1 31.5 9.7 8.5 10.4 10.9 11.5 12.3 7.2 Intermediate 11.3 15.6 15.0 13.1 37.5 9.6 12.8 12.1 9.3 10.7 12.6 5.8 Novice 8.9 16.4 11.5 13.6 29.2 22.9 10.1 10.7 8.3 8.8 10.2 13.1 S.D. 3.6 6.0 5.3 6.2 10.1 7.8 3.2 3.5 2.5 3.1 2.7 4.3 Appendix V Relationship Between Eye Movements and Introspection About Performance Body Segments I d e n t i f i e d by both Eye Movements and Introspection Walkover Handspring Unevens Beam-Walk Beam-Hand Vault % T o t a l 1 2 3 4 5 a 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 Congruence E l i t e . 6 1 3 4 2 2 1 2 1 1 5 3 2 1 2 1 3 1 68.3% D Inter-mediate 4 2 4 1 1 1 4 3 3 3 1 2 7 1 8 76.3% Novice 2 6 5 1 3 2 3 4 1 4 1 1 5 1 1 66.6% Average 70.6% Legend: 1-Head; 2-Shoulders; 3-Arms; 4-Hips; 5-Legs; Offbody i s not included as subjects were not asked to introspect about t h i s . k n = 10 for each group 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0076984/manifest

Comment

Related Items